def process_constituent_row(constituent, current_id):
constituent_id = row[indices['constituent_id_index']]
type_key = int(row[indices['type_id_index']])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related objects
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type, ELASTICSEARCH_INDEX):
constituent = elasticsearch_connection.get_item(constituent_id, type, ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % constituent_id)
return (constituent, current_id)
if 'altnames' not in constituent:
constituent['altnames'] = []
altname = row[indices['altname_index']]
name_type = row[indices['name_type_index']]
constituent['altnames'].append({
'name' : altname,
'type' : name_type
})
return (constituent, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices["constituent_id_index"]]
type_key = int(row[indices["type_id_index"]])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related published
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type):
constituent = elasticsearch_connection.get_item(constituent_id, type)
else:
print "%s could not be found!" % constituent_id
return (constituent, current_id)
if "relateditems" not in constituent:
constituent["relateditems"] = {}
reference_id = row[indices["reference_id_index"]]
title = row[indices["title_index"]]
boiler_text = row[indices["boiler_text_index"]]
date = row[indices["date_index"]]
main_url = get_media_url(row[indices["path_index"]], row[indices["file_index"]])
if "pubdocs" not in constituent["relateditems"]:
constituent["relateditems"]["pubdocs"] = []
constituent["relateditems"]["pubdocs"].append(
{"id": reference_id, "boilertext": boiler_text, "displaytext": title, "date": date, "url": main_url}
)
# keep the related items sorted
constituent["relateditems"]["pubdocs"].sort(key=operator.itemgetter("displaytext"))
return (constituent, current_id)
def process_media_row(media, current_id):
id = row[indices['id_index']]
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
# for now, ignore Microfilm and Document media types
if (media_type_key in [4, 5]):
return (media, current_id)
if id != current_id:
save(media)
current_id = id
media = {}
if elasticsearch_connection.item_exists(id, media_type):
media = elasticsearch_connection.get_item(id, media_type)
else:
print "%s could not be found!" % id
return (media, current_id)
if 'relateditems' not in media:
media['relateditems'] = {}
constituent_id = row[indices['constituent_id_index']]
display_name = row[indices['display_name_index']]
description = row[indices['remarks_index']] if row[
indices['remarks_index']] != "NULL" else ""
display_date = ""
if row[indices['display_date_index']] != "NULL":
display_date = row[indices['display_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
constituent_dict = {}
role = row[indices['role_index']]
# update the set of roles for this media
if role not in media['roles']:
# make sure Photographer is first
if role == "Photographer":
media['roles'].insert(0, role)
else:
media['roles'].append(role)
constituent_dict['role'] = row[indices['role_index']]
constituent_dict['roleid'] = row[indices['role_id_index']]
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_dict['description'] = description
constituent_dict['thumbnail'] = thumbnail_url
constituent_type_key = int(row[indices['constituent_type_id_index']])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
if constituent_type not in media['relateditems']:
media['relateditems'][constituent_type] = []
media['relateditems'][constituent_type].append(constituent_dict)
# keep the related items sorted
media['relateditems'][constituent_type].sort(
key=operator.itemgetter('displaytext'))
return (media, current_id)
def process_site_row(site, current_id):
site_id = row[site_id_index]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related photos
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return(site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
media_master_id = row[media_master_id_index]
if "photos" not in site['relateditems']:
site['relateditems']["photos"] = []
site['relateditems']["photos"].append({
'id' : media_master_id,
'displaytext' : media_master_id
})
return(site, current_id)
def process_pub_row(pub, current_id):
pub_id = row[indices['pub_id_index']]
if pub_id != current_id:
save(pub)
current_id = pub_id
pub = {}
if elasticsearch_connection.item_exists(pub_id, 'pubdocs',
ELASTICSEARCH_INDEX):
pub = elasticsearch_connection.get_item(
pub_id, 'pubdocs', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % pub_id)
return (pub, current_id)
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
main_url = get_media_url(row[indices['main_path_index']],
row[indices['main_file_index']])
pub['primarydisplay'] = {
'thumbnail': thumbnail_url,
'main': thumbnail_url
}
pub['pdf'] = main_url
return (pub, current_id)
def process_object_row(object, current_id):
object_id = row[indices['object_id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if object_id != current_id:
# will likely have multiple rows
save(object)
current_id = object_id
object = {}
if elasticsearch_connection.item_exists(object_id, classification):
object = elasticsearch_connection.get_item(
object_id, classification)
else:
print "%s could not be found!" % object_id
return (object, current_id)
if 'altnums' not in object:
object['altnums'] = []
altnum = row[indices['altnum_index']]
prefix_idx = altnum.find('_')
without_prefix = altnum[prefix_idx + 1:]
description = row[indices['description_index']] if row[
indices['description_index']] != "NULL" else ""
object['altnums'].append({
"altnum": altnum,
"description": description,
'without_prefix': without_prefix
})
object['allnumbers'].extend((altnum, without_prefix))
return (object, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related constituents
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return(site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
constituent_id = row[indices['constituent_id_index']]
display_name = row[indices['display_name_index']]
display_date = ""
if row[indices['display_date_index']] != "NULL":
display_date = row[indices['display_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
constituent_dict = {}
role = row[indices['role_index']]
# update the set of roles for this site
if role not in site['roles']:
# make sure Tomb Owner is first
if role == "Tomb Owner":
site['roles'].insert(0, role)
else:
site['roles'].append(role)
description = row[indices['remarks_index']] if row[indices['remarks_index']] != "NULL" else ""
constituent_dict['role'] = role
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_dict['description'] = description
constituent_dict['thumbnail'] = thumbnail_url
constituent_type_key = int(row[indices['constituent_type_id_index']])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
# add to array of people for easier searching
if (constituent_type_key in [1,3]):
site['people'].append(display_name)
if constituent_type not in site['relateditems']:
site['relateditems'][constituent_type] = []
site['relateditems'][constituent_type].append(constituent_dict)
# keep the related items sorted
site['relateditems'][constituent_type].sort(key=operator.itemgetter('displaytext'))
if role == 'Tomb Owner':
site['tombowner'] = "Yes"
return(site, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
if site_id != current_id:
# will likely have multiple rows for one site because of many related objects
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return (site, current_id)
if 'sitedates' not in site:
site['sitedates'] = []
event_type = row[indices['event_type_index']]
date_text = row[indices['date_text_index']]
site['sitedates'].append({
'type' : event_type,
'date' : date_text
})
site['datevalues'].append(date_text)
return (site, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related objects
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return (site, current_id)
if 'altnums' not in site:
site['altnums'] = []
if 'altnum_types' not in site:
site['altnum_types'] = []
altnum = row[indices['altnum_index']]
description = row[indices['description_index']] if row[indices['description_index']] != "NULL" else ""
if description not in site['altnum_types']:
site['altnum_types'].append(description)
site['altnums'].append({"altnum" : altnum, "description" : description})
return (site, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
if site_id != current_id:
# will likely have multiple rows for one site because of many related objects
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites',
ELASTICSEARCH_INDEX):
site = elasticsearch_connection.get_item(
site_id, 'sites', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % site_id)
return (site, current_id)
if 'sitedates' not in site:
site['sitedates'] = []
event_type = row[indices['event_type_index']]
date_text = row[indices['date_text_index']]
site['sitedates'].append({'type': event_type, 'date': date_text})
site['datevalues'].append(date_text)
return (site, current_id)
def process_object_row(object, current_id):
object_id = row[indices['object_id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if object_id != current_id:
# will likely have multiple rows
save(object)
current_id = object_id
object = {}
if elasticsearch_connection.item_exists(object_id, classification):
object = elasticsearch_connection.get_item(
object_id, classification)
else:
print "%s could not be found!" % object_id
return (object, current_id)
if 'flexfields' not in object:
object['flexfields'] = {}
groupname = row[indices['group_name_index']]
if groupname not in object['flexfields']:
object['flexfields'][groupname] = []
fieldname = row[indices['field_name_index']]
fieldvalue = row[indices['field_value_index']]
object['flexfields'][groupname].append({fieldname: fieldvalue})
return (object, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related objects
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites',
ELASTICSEARCH_INDEX):
site = elasticsearch_connection.get_item(
site_id, 'sites', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % site_id)
return (site, current_id)
if 'altnums' not in site:
site['altnums'] = []
if 'altnum_types' not in site:
site['altnum_types'] = []
altnum = row[indices['altnum_index']]
description = row[indices['description_index']] if row[
indices['description_index']] != "NULL" else ""
if description not in site['altnum_types']:
site['altnum_types'].append(description)
site['altnums'].append({"altnum": altnum, "description": description})
return (site, current_id)
def process_object_row(object, current_id):
object_id = row[indices['object_id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if object_id != current_id:
# will likely have multiple rows
save(object)
current_id = object_id
object = {}
if elasticsearch_connection.item_exists(object_id, classification):
object = elasticsearch_connection.get_item(object_id, classification)
else:
print "%s could not be found!" % object_id
return (object, current_id)
if 'altnums' not in object:
object['altnums'] = []
altnum = row[indices['altnum_index']]
prefix_idx = altnum.find('_')
without_prefix = altnum[prefix_idx+1:]
description = row[indices['description_index']] if row[indices['description_index']] != "NULL" else ""
object['altnums'].append({"altnum" : altnum, "description" : description, 'without_prefix': without_prefix})
object['allnumbers'].extend((altnum, without_prefix))
return (object, current_id)
def process_object_row(object, current_id):
object_id = row[indices['object_id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if object_id != current_id:
# will likely have multiple rows
save(object)
current_id = object_id
object = {}
if elasticsearch_connection.item_exists(object_id, classification):
object = elasticsearch_connection.get_item(object_id, classification)
else:
print "%s could not be found!" % object_id
return (object, current_id)
if 'flexfields' not in object:
object['flexfields'] = {}
groupname = row[indices['group_name_index']]
if groupname not in object['flexfields']:
object['flexfields'][groupname] = []
fieldname = row[indices['field_name_index']]
fieldvalue = row[indices['field_value_index']]
object['flexfields'][groupname].append({fieldname : fieldvalue})
return (object, current_id)
def process_pub_row(pub, current_id):
id = row[indices['id_index']]
if id != current_id:
save(pub)
current_id = id
pub = {}
if elasticsearch_connection.item_exists(id, 'pubdocs'):
pub = elasticsearch_connection.get_item(id, 'pubdocs')
else:
print "%s could not be found!" % id
return(pub, current_id)
if 'relateditems' not in pub:
pub['relateditems'] = {}
site_id = row[indices['site_id_index']]
site_name = row[indices['site_name_index']]
site_number = row[indices['site_number_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = site_number
site_dict['thumbnail'] = thumbnail_url
if 'sites' not in pub['relateditems']:
pub['relateditems']['sites'] = []
pub['relateditems']['sites'].append(site_dict)
# keep the related items sorted
pub['relateditems']['sites'].sort(key=operator.itemgetter('displaytext'))
return(pub, current_id)
def process_object_row(object, current_id):
id = row[id_index]
classification_key = int(row[classification_id_index])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
media_master_id = row[media_master_id_index]
if 'photos' not in object['relateditems']:
object['relateditems']['photos'] = []
object['relateditems']['photos'].append({
'id' : media_master_id,
'displaytext' : media_master_id})
return(object, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices['constituent_id_index']]
type_key = int(row[indices['type_id_index']])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related objects
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type):
constituent = elasticsearch_connection.get_item(
constituent_id, type)
else:
print "%s could not be found!" % constituent_id
return (constituent, current_id)
if 'relateditems' not in constituent:
constituent['relateditems'] = {}
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[indices['object_id_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
date = "" if row[indices['object_date_index']].lower(
) == "null" else row[indices['object_date_index']]
object_title = row[indices['object_title_index']]
object_number = row[indices['object_number_index']]
if classification == "diarypages" and object_title.lower() == "null":
idx = object_number.find('_')
object_title = object_number[idx + 1:]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in constituent['relateditems']:
constituent['relateditems'][classification] = []
constituent['relateditems'][classification].append({
'id':
object_id,
'title':
object_title,
'displaytext':
object_title,
'classificationid':
classification_key,
'number':
object_number,
'date':
date,
'thumbnail':
thumbnail_url
})
# keep the related items sorted
constituent['relateditems'][classification].sort(
key=operator.itemgetter('displaytext'))
return (constituent, current_id)
def process_pub_row(pub, current_id):
pub_id = row[indices['pub_id_index']]
if pub_id != current_id:
# will likely have multiple rows for one pub because of many related constituents
# only get a new pub if we have a new pub id, but first save old pub to elasticsearch
save(pub)
current_id = pub_id
pub = {}
if elasticsearch_connection.item_exists(pub_id, 'pubdocs'):
pub = elasticsearch_connection.get_item(pub_id, 'pubdocs')
else:
print "%s could not be found!" % pub_id
return (pub, current_id)
if 'relateditems' not in pub:
pub['relateditems'] = {}
constituent_id = row[indices['constituent_id_index']]
display_name = row[indices['display_name_index']]
display_date = ""
if row[indices['display_date_index']] != "NULL":
display_date = row[indices['display_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
alpha_sort = row[indices['alpha_sort_index']]
constituent_dict = {}
role = row[indices['role_index']]
# update the set of roles for this pub
if role not in pub['roles']:
pub['roles'].append(role)
if role == "Author":
pub["authors"].append(alpha_sort)
description = row[indices['remarks_index']] if row[
indices['remarks_index']] != "NULL" else ""
constituent_dict['role'] = role
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_dict['description'] = description
constituent_dict['thumbnail'] = thumbnail_url
constituent_type_key = int(row[indices['constituent_type_id_index']])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
if constituent_type not in pub['relateditems']:
pub['relateditems'][constituent_type] = []
pub['relateditems'][constituent_type].append(constituent_dict)
# keep the related items sorted
pub['relateditems'][constituent_type].sort(
key=operator.itemgetter('displaytext'))
return (pub, current_id)
def process_pub_row(pub, current_id):
pub_id = row[indices['pub_id_index']]
if pub_id != current_id:
save(pub)
current_id = pub_id
pub = {}
if elasticsearch_connection.item_exists(pub_id, 'pubdocs',
ELASTICSEARCH_INDEX):
pub = elasticsearch_connection.get_item(
pub_id, 'pubdocs', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % pub_id)
return (pub, current_id)
if 'relateditems' not in pub:
pub['relateditems'] = {}
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[indices['object_id_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
drs_id = "" if row[indices['drs_id']].lower() == "null" else row[
indices['drs_id']]
has_manifest = False if drs_id == "" else True
if not thumbnail_url and drs_id:
thumbnail_url = create_thumbnail_url(drs_id)
date = "" if row[indices['object_date_index']].lower(
) == "null" else row[indices['object_date_index']]
object_title = row[indices['object_title_index']]
object_number = row[indices['object_number_index']]
if classification == "diarypages" and object_title.lower() == "null":
idx = object_number.find('_')
object_title = object_number[idx + 1:]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in pub['relateditems']:
pub['relateditems'][classification] = []
pub['relateditems'][classification].append({
'id': object_id,
'title': object_title,
'displaytext': object_title,
'classificationid': classification_key,
'number': object_number,
'date': date,
'thumbnail': thumbnail_url,
'has_manifest': has_manifest
})
# keep the related items sorted
pub['relateditems'][classification].sort(
key=operator.itemgetter('displaytext'))
return (pub, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return (object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
constituent_id = row[indices['constituent_id_index']]
display_name = row[indices['display_name_index']]
description = row[indices['remarks_index']] if row[
indices['remarks_index']] != "NULL" else ""
display_date = ""
if row[indices['display_date_index']] != "NULL":
display_date = row[indices['display_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
constituent_dict = {}
role = row[indices['role_index']]
# update the set of roles for this object
if role not in object['roles']:
object['roles'].append(role)
constituent_dict['role'] = row[indices['role_index']]
constituent_dict['roleid'] = row[indices['role_id_index']]
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_dict['description'] = description
constituent_dict['thumbnail'] = thumbnail_url
constituent_type_key = int(row[indices['constituent_type_id_index']])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
if constituent_type not in object['relateditems']:
object['relateditems'][constituent_type] = []
object['relateditems'][constituent_type].append(constituent_dict)
# keep the related items sorted
object['relateditems'][constituent_type].sort(
key=operator.itemgetter('displaytext'))
return (object, current_id)
def process_object_row(object, current_id):
id = row[id_index]
classification_key = int(row[classification_id_index])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
constituent_id = row[constituent_id_index]
display_name = row[display_name_index]
display_date = ""
if row[display_date_index] != "NULL":
display_date = row[display_date_index]
constituent_dict = {}
constituent_dict['role'] = row[role_index]
constituent_dict['roleid'] = row[role_id_index]
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_type_key = int(row[constituent_type_id_index])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
if constituent_type not in object['relateditems']:
object['relateditems'][constituent_type] = []
object['relateditems'][constituent_type].append(constituent_dict)
# parse out any constituents "Mentioned on this page" (RoleID==48)
if constituent_dict['roleid'] == '48':
if 'mentioned' not in object:
object['mentioned'] = {}
if 'people' not in object['mentioned']:
object['mentioned']['people'] = []
object['mentioned']['people'].append(constituent_dict)
# parse out any "Author" (RoleID==50)
if constituent_dict['roleid'] == '50':
if 'author' not in object:
object['author'] = []
object['author'].append(constituent_dict)
return(object, current_id)
def process_media_row(media, current_id):
id = row[indices['media_id_index']]
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
# for now, ignore Microfilm and Document media types
if (media_type_key in [4, 5]):
return (media, current_id)
if id != current_id:
save(media)
current_id = id
media = {}
if elasticsearch_connection.item_exists(id, media_type):
media = elasticsearch_connection.get_item(id, media_type)
else:
print "%s could not be found!" % id
return (media, current_id)
if 'relateditems' not in media:
media['relateditems'] = {}
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[indices['object_id_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
date = "" if row[indices['object_date_index']].lower(
) == "null" else row[indices['object_date_index']]
object_title = row[indices['object_title_index']]
object_number = row[indices['object_number_index']]
if classification == "diarypages" and object_title.lower() == "null":
idx = object_number.find('_')
object_title = object_number[idx + 1:]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in media['relateditems']:
media['relateditems'][classification] = []
media['relateditems'][classification].append({
'id': object_id,
'title': object_title,
'displaytext': object_title,
'classificationid': classification_key,
'number': object_number,
'date': date,
'thumbnail': thumbnail_url
})
# keep the related items sorted
media['relateditems'][classification].sort(
key=operator.itemgetter('displaytext'))
return (media, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
media_master_id = row[indices['media_master_id_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
main_url = get_media_url(row[indices['main_path_index']], row[indices['main_file_index']])
display_text = row[indices['caption_index']]
# this is a bit of a hack because the MediaFormats for videos (in the TMS database) does not correctly identify the type of video
# so, make sure we are only using videos that are mp4s
if media_type_key == 3:
if not row[indices['main_file_index']].endswith('mp4'):
return(object, current_id)
if media_type not in object['relateditems']:
object['relateditems'][media_type] = []
if media_type == 'photos':
object['hasphoto'] = "Yes"
# add primary photo as a top level item as well
if row[indices['primary_display_index']] == '1':
object['primarydisplay'] = {
'thumbnail' : thumbnail_url,
'main' : main_url,
'displaytext' : display_text
}
if not (classification == '3dmodels' and media_type == '3dmodels'):
object['relateditems'][media_type].append({
'id' : media_master_id,
'displaytext' : display_text,
'primarydisplay' : True if row[indices['primary_display_index']] == '1' else False,
'thumbnail' : thumbnail_url,
'main' : main_url
})
return(object, current_id)
def process_pub_row(pub, current_id):
pub_id = row[indices['pub_id_index']]
if pub_id != current_id:
# will likely have multiple rows for one pub because of many related constituents
# only get a new pub if we have a new pub id, but first save old pub to elasticsearch
save(pub)
current_id = pub_id
pub = {}
if elasticsearch_connection.item_exists(pub_id, 'pubdocs'):
pub = elasticsearch_connection.get_item(pub_id, 'pubdocs')
else:
print "%s could not be found!" % pub_id
return(pub, current_id)
if 'relateditems' not in pub:
pub['relateditems'] = {}
constituent_id = row[indices['constituent_id_index']]
display_name = row[indices['display_name_index']]
display_date = ""
if row[indices['display_date_index']] != "NULL":
display_date = row[indices['display_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
alpha_sort = row[indices['alpha_sort_index']]
constituent_dict = {}
role = row[indices['role_index']]
# update the set of roles for this pub
if role not in pub['roles']:
pub['roles'].append(role)
if role == "Author":
pub["authors"].append(alpha_sort)
description = row[indices['remarks_index']] if row[indices['remarks_index']] != "NULL" else ""
constituent_dict['role'] = role
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_dict['description'] = description
constituent_dict['thumbnail'] = thumbnail_url
constituent_type_key = int(row[indices['constituent_type_id_index']])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
if constituent_type not in pub['relateditems']:
pub['relateditems'][constituent_type] = []
pub['relateditems'][constituent_type].append(constituent_dict)
# keep the related items sorted
pub['relateditems'][constituent_type].sort(key=operator.itemgetter('displaytext'))
return(pub, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
constituent_id = row[indices['constituent_id_index']]
display_name = row[indices['display_name_index']]
description = row[indices['remarks_index']] if row[indices['remarks_index']] != "NULL" else ""
display_date = ""
if row[indices['display_date_index']] != "NULL":
display_date = row[indices['display_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
constituent_dict = {}
role = row[indices['role_index']]
# update the set of roles for this object
if role not in object['roles']:
object['roles'].append(role)
constituent_dict['role'] = row[indices['role_index']]
constituent_dict['roleid'] = row[indices['role_id_index']]
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_dict['description'] = description
constituent_dict['thumbnail'] = thumbnail_url
constituent_type_key = int(row[indices['constituent_type_id_index']])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
if constituent_type not in object['relateditems']:
object['relateditems'][constituent_type] = []
object['relateditems'][constituent_type].append(constituent_dict)
# keep the related items sorted
object['relateditems'][constituent_type].sort(key=operator.itemgetter('displaytext'))
return(object, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices["constituent_id_index"]]
type_key = int(row[indices["type_id_index"]])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related objects
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type):
constituent = elasticsearch_connection.get_item(constituent_id, type)
else:
print "%s could not be found!" % constituent_id
return (constituent, current_id)
if "relateditems" not in constituent:
constituent["relateditems"] = {}
classification_key = int(row[indices["classification_id_index"]])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[indices["object_id_index"]])
thumbnail_url = get_media_url(row[indices["thumb_path_index"]], row[indices["thumb_file_index"]])
date = "" if row[indices["object_date_index"]].lower() == "null" else row[indices["object_date_index"]]
object_title = row[indices["object_title_index"]]
object_number = row[indices["object_number_index"]]
if classification == "diarypages" and object_title.lower() == "null":
idx = object_number.find("_")
object_title = object_number[idx + 1 :]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in constituent["relateditems"]:
constituent["relateditems"][classification] = []
constituent["relateditems"][classification].append(
{
"id": object_id,
"title": object_title,
"displaytext": object_title,
"classificationid": classification_key,
"number": object_number,
"date": date,
"thumbnail": thumbnail_url,
}
)
# keep the related items sorted
constituent["relateditems"][classification].sort(key=operator.itemgetter("displaytext"))
return (constituent, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related objects
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return (site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[indices['object_id_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
date = "" if row[indices['object_date_index']].lower(
) == "null" else row[indices['object_date_index']]
object_title = row[indices['object_title_index']]
object_number = row[indices['object_number_index']]
if classification == "diarypages" and object_title.lower() == "null":
idx = object_number.find('_')
object_title = object_number[idx + 1:]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in site['relateditems']:
site['relateditems'][classification] = []
site['relateditems'][classification].append({
'id': object_id,
'title': object_title,
'displaytext': object_title,
'classificationid': classification_key,
'number': object_number,
'date': date,
'thumbnail': thumbnail_url
})
# keep the related items sorted
site['relateditems'][classification].sort(
key=operator.itemgetter('displaytext'))
return (site, current_id)
def process_media_row(media, current_id):
id = row[indices['id_index']]
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
# for now, ignore Microfilm and Document media types
if (media_type_key in [4, 5]):
return (media, current_id)
if id != current_id:
save(media)
current_id = id
media = {}
if elasticsearch_connection.item_exists(id, media_type,
ELASTICSEARCH_INDEX):
media = elasticsearch_connection.get_item(
id, media_type, ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % id)
return (media, current_id)
if 'relateditems' not in media:
media['relateditems'] = {}
site_id = row[indices['site_id_index']]
site_name = row[indices['site_name_index']]
site_number = row[indices['site_number_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
drs_id = "" if row[indices['drs_id']].lower() == "null" else row[
indices['drs_id']]
if not thumbnail_url and drs_id:
thumbnail_url = create_thumbnail_url(drs_id)
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = site_number
site_dict['thumbnail'] = thumbnail_url
if 'sites' not in media['relateditems']:
media['relateditems']['sites'] = []
media['relateditems']['sites'].append(site_dict)
# keep the related items sorted
media['relateditems']['sites'].sort(
key=operator.itemgetter('displaytext'))
return (media, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related photos
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return(site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
media_master_id = row[indices['media_master_id_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
main_url = get_media_url(row[indices['main_path_index']], row[indices['main_file_index']])
display_text = row[indices['caption_index']]
# this is a bit of a hack because the MediaFormats for videos (in the TMS database) does not correctly identify the type of video
# so, make sure we are only using videos that are mp4s
if media_type_key == 3:
if not row[indices['main_file_index']].endswith('mp4'):
return(site, current_id)
if media_type not in site['relateditems']:
site['relateditems'][media_type] = []
# add primary photo as a top level item as well
if row[indices['primary_display_index']] == '1':
site['primarydisplay'] = {
'thumbnail' : thumbnail_url,
'main' : main_url,
'displaytext' : display_text
}
site['relateditems'][media_type].append({
'id' : media_master_id,
'displaytext' : display_text,
'primarydisplay' : True if row[indices['primary_display_index']] == '1' else False,
'thumbnail' : thumbnail_url,
'main' : main_url
})
return(site, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return (object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
site_id = row[indices['site_id_index']]
site_name = row[indices['site_name_index']]
site_number = row[indices['site_number_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = site_number
site_dict['thumbnail'] = thumbnail_url
if 'sites' not in object['relateditems']:
object['relateditems']['sites'] = []
object['relateditems']['sites'].append(site_dict)
# keep the related items sorted
object['relateditems']['sites'].sort(
key=operator.itemgetter('displaytext'))
# for unpubdocs, add sites for "Mentioned on this page"
if classification == "unpubdocs":
if 'mentioned' not in object:
object['mentioned'] = {}
if 'sites' not in object['mentioned']:
object['mentioned']['sites'] = []
object['mentioned']['sites'].append(site_dict)
return (object, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices["constituent_id_index"]]
type_key = int(row[indices["type_id_index"]])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related photos
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type):
constituent = elasticsearch_connection.get_item(constituent_id, type)
else:
print "%s could not be found!" % constituent_id
return (constituent, current_id)
if "relateditems" not in constituent:
constituent["relateditems"] = {}
media_type_key = int(row[indices["media_type_id_index"]])
media_type = MEDIATYPES.get(media_type_key)
media_master_id = row[indices["media_master_id_index"]]
thumbnail_url = get_media_url(row[indices["thumb_path_index"]], row[indices["thumb_file_index"]])
main_url = get_media_url(row[indices["main_path_index"]], row[indices["main_file_index"]])
display_text = row[indices["caption_index"]]
# this is a bit of a hack because the MediaFormats for videos (in the TMS database) does not correctly identify the type of video
# so, make sure we are only using videos that are mp4s
if media_type_key == 3:
if not row[indices["main_file_index"]].endswith("mp4"):
return (constituent, current_id)
if media_type not in constituent["relateditems"]:
constituent["relateditems"][media_type] = []
# add primary photo as a top level item as well
if row[indices["primary_display_index"]] == "1":
constituent["primarydisplay"] = {"thumbnail": thumbnail_url, "main": main_url, "displaytext": display_text}
constituent["relateditems"][media_type].append(
{
"id": media_master_id,
"displaytext": display_text,
"primarydisplay": True if row[indices["primary_display_index"]] == "1" else False,
"thumbnail": thumbnail_url,
"main": main_url,
}
)
return (constituent, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
site_id = row[indices['site_id_index']]
site_name = row[indices['site_name_index']]
site_number = row[indices['site_number_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = site_number
site_dict['thumbnail'] = thumbnail_url
if 'sites' not in object['relateditems']:
object['relateditems']['sites'] = []
object['relateditems']['sites'].append(site_dict)
# keep the related items sorted
object['relateditems']['sites'].sort(key=operator.itemgetter('displaytext'))
# for unpubdocs, add sites for "Mentioned on this page"
if classification == "unpubdocs":
if 'mentioned' not in object:
object['mentioned'] = {}
if 'sites' not in object['mentioned']:
object['mentioned']['sites'] = []
object['mentioned']['sites'].append(site_dict)
return(object, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related objects
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return (site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[indices['object_id_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
date = "" if row[indices['object_date_index']].lower() == "null" else row[indices['object_date_index']]
object_title = row[indices['object_title_index']]
object_number = row[indices['object_number_index']]
if classification == "diarypages" and object_title.lower() == "null":
idx = object_number.find('_')
object_title = object_number[idx+1:]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in site['relateditems']:
site['relateditems'][classification] = []
site['relateditems'][classification].append({
'id' : object_id,
'title' : object_title,
'displaytext' : object_title,
'classificationid' : classification_key,
'number' : object_number,
'date' : date,
'thumbnail' : thumbnail_url})
# keep the related items sorted
site['relateditems'][classification].sort(key=operator.itemgetter('displaytext'))
return (site, current_id)
def process_pub_row(pub, current_id):
id = row[indices['id_index']]
if id != current_id:
save(pub)
current_id = id
pub = {}
if elasticsearch_connection.item_exists(id, 'pubdocs',
ELASTICSEARCH_INDEX):
pub = elasticsearch_connection.get_item(
id, 'pubdocs', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % id)
return (pub, current_id)
if 'relateditems' not in pub:
pub['relateditems'] = {}
site_id = row[indices['site_id_index']]
site_name = row[indices['site_name_index']]
site_number = row[indices['site_number_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
drs_id = "" if row[indices['drs_id']].lower() == "null" else row[
indices['drs_id']]
has_manifest = False if drs_id == "" else True
if not thumbnail_url and drs_id:
thumbnail_url = create_thumbnail_url(drs_id)
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = site_number
site_dict['thumbnail'] = thumbnail_url
site_dict['has_manifest'] = has_manifest
if 'sites' not in pub['relateditems']:
pub['relateditems']['sites'] = []
pub['relateditems']['sites'].append(site_dict)
# keep the related items sorted
pub['relateditems']['sites'].sort(
key=operator.itemgetter('displaytext'))
return (pub, current_id)
def process_media_row(media, current_id):
id = row[indices['id_index']]
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
# for now, ignore Microfilm and Document media types
if (media_type_key in [4, 5]):
return (media, current_id)
if id != current_id:
save(media)
current_id = id
media = {}
if elasticsearch_connection.item_exists(id, media_type,
ELASTICSEARCH_INDEX):
media = elasticsearch_connection.get_item(
id, media_type, ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % id)
return (media, current_id)
if 'relateditems' not in media:
media['relateditems'] = {}
reference_id = row[indices['reference_id_index']]
title = row[indices['title_index']]
boiler_text = row[indices['boiler_text_index']]
date = row[indices['date_index']]
main_url = get_media_url(row[indices['path_index']],
row[indices['file_index']])
if 'pubdocs' not in media['relateditems']:
media['relateditems']['pubdocs'] = []
media['relateditems']['pubdocs'].append({
'id': reference_id,
'boilertext': boiler_text,
'displaytext': boiler_text,
'date': date,
'url': main_url
})
# keep the related items sorted
media['relateditems']['pubdocs'].sort(
key=operator.itemgetter('displaytext'))
return (media, current_id)
def process_pub_row(pub, current_id):
pub_id = row[indices['pub_id_index']]
if pub_id != current_id:
save(pub)
current_id = pub_id
pub = {}
if elasticsearch_connection.item_exists(pub_id, 'pubdocs'):
pub = elasticsearch_connection.get_item(pub_id, 'pubdocs')
else:
print "%s could not be found!" % pub_id
return (pub, current_id)
if 'relateditems' not in pub:
pub['relateditems'] = {}
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[indices['object_id_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
date = "" if row[indices['object_date_index']].lower() == "null" else row[indices['object_date_index']]
object_title = row[indices['object_title_index']]
object_number = row[indices['object_number_index']]
if classification == "diarypages" and object_title.lower() == "null":
idx = object_number.find('_')
object_title = object_number[idx+1:]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in pub['relateditems']:
pub['relateditems'][classification] = []
pub['relateditems'][classification].append({
'id' : object_id,
'title' : object_title,
'displaytext' : object_title,
'classificationid' : classification_key,
'number' : object_number,
'date' : date,
'thumbnail' : thumbnail_url})
# keep the related items sorted
pub['relateditems'][classification].sort(key=operator.itemgetter('displaytext'))
return (pub, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices['constituent_id_index']]
type_key = int(row[indices['type_id_index']])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related constituents
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type, ELASTICSEARCH_INDEX):
constituent = elasticsearch_connection.get_item(constituent_id, type, ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % constituent_id)
return(constituent, current_id)
if 'relateditems' not in constituent:
constituent['relateditems'] = {}
site_id = row[indices['site_id_index']]
site_name = row[indices['site_name_index']]
site_number = row[indices['site_number_index']]
drs_id = "" if row[indices['drs_id']].lower() == "null" else row[indices['drs_id']]
has_manifest = False if drs_id == "" else True
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
if not thumbnail_url and drs_id:
thumbnail_url = create_thumbnail_url(drs_id)
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = "%s, %s" % (site_name, site_number)
site_dict['thumbnail'] = thumbnail_url
site_dict['has_manifest'] = has_manifest
if 'sites' not in constituent['relateditems']:
constituent['relateditems']['sites'] = []
constituent['relateditems']['sites'].append(site_dict)
# keep the related items sorted
constituent['relateditems']['sites'].sort(key=operator.itemgetter('displaytext'))
return(constituent, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related published
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites',
ELASTICSEARCH_INDEX):
site = elasticsearch_connection.get_item(
site_id, 'sites', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % site_id)
return (site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
reference_id = row[indices['reference_id_index']]
title = row[indices['title_index']]
boiler_text = row[indices['boiler_text_index']]
date = row[indices['date_index']]
main_url = get_media_url(row[indices['path_index']],
row[indices['file_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
if "pubdocs" not in site['relateditems']:
site['relateditems']["pubdocs"] = []
site['relateditems']["pubdocs"].append({
'id': reference_id,
'boilertext': boiler_text,
'displaytext': boiler_text,
'date': date,
'url': main_url,
'thumbnail': thumbnail_url
})
# keep the related items sorted
site['relateditems']['pubdocs'].sort(
key=operator.itemgetter('displaytext'))
return (site, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices['constituent_id_index']]
type_key = int(row[indices['type_id_index']])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related published
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type):
constituent = elasticsearch_connection.get_item(
constituent_id, type)
else:
print "%s could not be found!" % constituent_id
return (constituent, current_id)
if 'relateditems' not in constituent:
constituent['relateditems'] = {}
reference_id = row[indices['reference_id_index']]
title = row[indices['title_index']]
boiler_text = row[indices['boiler_text_index']]
date = "" if row[indices['date_index']].lower() == "null" else row[
indices['date_index']]
main_url = get_media_url(row[indices['path_index']],
row[indices['file_index']])
if "pubdocs" not in constituent['relateditems']:
constituent['relateditems']["pubdocs"] = []
constituent['relateditems']["pubdocs"].append({
'id': reference_id,
'boilertext': boiler_text,
'displaytext': title,
'date': date,
'url': main_url
})
# keep the related items sorted
constituent['relateditems']['pubdocs'].sort(
key=operator.itemgetter('displaytext'))
return (constituent, current_id)
def process_object_row(object, current_id):
id = row[id_index]
classification_key = int(row[classification_id_index])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
site_id = row[site_id_index]
site_name = row[site_name_index]
site_number = row[site_number_index]
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = "%s, %s" % (site_name, site_number)
if 'sites' not in object['relateditems']:
object['relateditems']['sites'] = []
object['relateditems']['sites'].append(site_dict)
# for unpubdocs, add sites for "Mentioned on this page"
if classification == "unpubdocs":
if 'mentioned' not in object:
object['mentioned'] = {}
if 'sites' not in object['mentioned']:
object['mentioned']['sites'] = []
object['mentioned']['sites'].append(site_dict)
return(object, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return (object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
unpublished_id = row[indices['unpublished_id_index']]
unpublished_title = row[indices['unpublished_title_index']]
number = row[indices['object_number_index']]
date = "" if row[indices['object_date_index']].lower(
) == "null" else row[indices['object_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
if 'unpubdocs' not in object['relateditems']:
object['relateditems']['unpubdocs'] = []
object['relateditems']['unpubdocs'].append({
'id': unpublished_id,
'text': unpublished_title,
'displaytext': unpublished_title,
'date': date,
'number': number,
'thumbnail': thumbnail_url
})
# keep the related items sorted
object['relateditems']['unpubdocs'].sort(
key=operator.itemgetter('displaytext'))
return (object, current_id)
def process_site_row(site, current_id):
site_id = row[site_id_index]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related constituents
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return(site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
constituent_id = row[constituent_id_index]
display_name = row[display_name_index]
display_date = ""
if row[display_date_index] != "NULL":
display_date = row[display_date_index]
constituent_dict = {}
role = row[role_index]
constituent_dict['role'] = role
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_type_key = int(row[constituent_type_id_index])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
if constituent_type not in site['relateditems']:
site['relateditems'][constituent_type] = []
site['relateditems'][constituent_type].append(constituent_dict)
if role == 'Tomb Owner':
site['tombowner'] = True
return(site, current_id)
def process_media_row(manifest, current_id):
media_id = row[indices['id_index']]
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
manifest_id = media_type + '-' + media_id
# ignore non-photos, if they somehow managed to show up in the query results
if media_type_key != 1:
return (manifest, current_id)
if manifest_id != current_id:
save(manifest)
current_id = manifest_id
manifest = {}
if elasticsearch_connection.item_exists(manifest_id, 'manifest',
ELASTICSEARCH_INDEX):
manifest = elasticsearch_connection.get_item(
manifest_id, 'manifest', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % manifest_id)
return (manifest, current_id)
role = "" if row[indices['role_index']].lower() == "null" else row[
indices['role_index']]
display_name = "" if row[indices['display_name_index']].lower(
) == "null" else row[indices['display_name_index']]
display_date = "" if row[indices['display_date_index']].lower(
) == "null" else row[indices['display_date_index']]
if display_date:
value = display_name + ", " + display_date
else:
value = display_name
manifest['manifest']['metadata'].append({
'label': role,
'value': value
})
return (manifest, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return (object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
reference_id = row[indices['reference_id_index']]
title = row[indices['title_index']]
boiler_text = row[indices['boiler_text_index']]
date = row[indices['date_index']]
main_url = get_media_url(row[indices['path_index']],
row[indices['file_index']])
if 'pubdocs' not in object['relateditems']:
object['relateditems']['pubdocs'] = []
object['relateditems']['pubdocs'].append({
'id': reference_id,
'boilertext': boiler_text,
'displaytext': boiler_text,
'date': date,
'url': main_url
})
# keep the related items sorted
object['relateditems']['pubdocs'].sort(
key=operator.itemgetter('displaytext'))
return (object, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices["constituent_id_index"]]
type_key = int(row[indices["type_id_index"]])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related constituents
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type):
constituent = elasticsearch_connection.get_item(constituent_id, type)
else:
print "%s could not be found!" % constituent_id
return (constituent, current_id)
if "relateditems" not in constituent:
constituent["relateditems"] = {}
site_id = row[indices["site_id_index"]]
site_name = row[indices["site_name_index"]]
site_number = row[indices["site_number_index"]]
thumbnail_url = get_media_url(row[indices["thumb_path_index"]], row[indices["thumb_file_index"]])
site_dict = {}
site_dict["id"] = site_id
site_dict["sitename"] = site_name
site_dict["sitenumber"] = site_number
site_dict["displaytext"] = "%s, %s" % (site_name, site_number)
site_dict["thumbnail"] = thumbnail_url
if "sites" not in constituent["relateditems"]:
constituent["relateditems"]["sites"] = []
constituent["relateditems"]["sites"].append(site_dict)
# keep the related items sorted
constituent["relateditems"]["sites"].sort(key=operator.itemgetter("displaytext"))
return (constituent, current_id)
def process_site_row(site, current_id):
site_id = row[site_id_index]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related objects
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return (site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
classification_key = int(row[classification_id_index])
classification = CLASSIFICATIONS.get(classification_key)
object_id = int(row[object_id_index])
object_title = row[object_title_index]
if classification == "diarypages" and object_title.lower() == "null":
object_number = row[object_number_index]
idx = object_number.find('_')
object_title = object_number[idx+1:]
if object_title.lower() == "null":
object_title = "[No Title]"
if classification not in site['relateditems']:
site['relateditems'][classification] = []
site['relateditems'][classification].append({
'id' : object_id,
'title' : object_title,
'displaytext' : object_title,
'classificationid' : classification_key})
return (site, current_id)
def process_constituent_row(constituent, current_id):
constituent_id = row[indices["constituent_id_index"]]
type_key = int(row[indices["type_id_index"]])
type = CONSTITUENTTYPES.get(type_key)
if constituent_id != current_id:
# will likely have multiple rows for one constituent because of many related objects
# only get a new constituent if we have a new constituent id, but first save old constituent to elasticsearch
save(constituent)
current_id = constituent_id
constituent = {}
if elasticsearch_connection.item_exists(constituent_id, type):
constituent = elasticsearch_connection.get_item(constituent_id, type)
else:
print "%s could not be found!" % constituent_id
return (constituent, current_id)
if "altnames" not in constituent:
constituent["altnames"] = []
altname = row[indices["altname_index"]]
name_type = row[indices["name_type_index"]]
constituent["altnames"].append({"name": altname, "type": name_type})
return (constituent, current_id)
def process_pub_row(pub, current_id):
pub_id = row[indices['pub_id_index']]
if pub_id != current_id:
save(pub)
current_id = pub_id
pub = {}
if elasticsearch_connection.item_exists(pub_id, 'pubdocs'):
pub = elasticsearch_connection.get_item(pub_id, 'pubdocs')
else:
print "%s could not be found!" % pub_id
return(pub, current_id)
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
main_url = get_media_url(row[indices['main_path_index']], row[indices['main_file_index']])
pub['primarydisplay'] = {
'thumbnail' : thumbnail_url,
'main' : thumbnail_url
}
pub['pdf'] = main_url
return(pub, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
unpublished_id = row[indices['unpublished_id_index']]
unpublished_title = row[indices['unpublished_title_index']]
number = row[indices['object_number_index']]
date = "" if row[indices['object_date_index']].lower() == "null" else row[indices['object_date_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
if 'unpubdocs' not in object['relateditems']:
object['relateditems']['unpubdocs'] = []
object['relateditems']['unpubdocs'].append({
'id' : unpublished_id,
'text' : unpublished_title,
'displaytext' : unpublished_title,
'date' : date,
'number' : number,
'thumbnail' : thumbnail_url})
# keep the related items sorted
object['relateditems']['unpubdocs'].sort(key=operator.itemgetter('displaytext'))
return(object, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related published
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return(site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
reference_id = row[indices['reference_id_index']]
title = row[indices['title_index']]
boiler_text = row[indices['boiler_text_index']]
date = row[indices['date_index']]
main_url = get_media_url(row[indices['path_index']], row[indices['file_index']])
thumbnail_url = get_media_url(row[indices['thumb_path_index']], row[indices['thumb_file_index']])
if "pubdocs" not in site['relateditems']:
site['relateditems']["pubdocs"] = []
site['relateditems']["pubdocs"].append({
'id' : reference_id,
'boilertext' : boiler_text,
'displaytext' : title,
'date' : date,
'url' : main_url,
'thumbnail' : thumbnail_url})
# keep the related items sorted
site['relateditems']['pubdocs'].sort(key=operator.itemgetter('displaytext'))
return(site, current_id)
def process_pub_row(pub, current_id):
id = row[indices['id_index']]
if id != current_id:
save(pub)
current_id = id
pub = {}
if elasticsearch_connection.item_exists(id, 'pubdocs'):
pub = elasticsearch_connection.get_item(id, 'pubdocs')
else:
print "%s could not be found!" % id
return (pub, current_id)
if 'relateditems' not in pub:
pub['relateditems'] = {}
site_id = row[indices['site_id_index']]
site_name = row[indices['site_name_index']]
site_number = row[indices['site_number_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
site_dict = {}
site_dict['id'] = site_id
site_dict['sitename'] = site_name
site_dict['sitenumber'] = site_number
site_dict['displaytext'] = site_number
site_dict['thumbnail'] = thumbnail_url
if 'sites' not in pub['relateditems']:
pub['relateditems']['sites'] = []
pub['relateditems']['sites'].append(site_dict)
# keep the related items sorted
pub['relateditems']['sites'].sort(
key=operator.itemgetter('displaytext'))
return (pub, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return (object, current_id)
geocode_dict = {}
geocode_dict['id'] = row[indices['geo_code_id_index']]
geocode_dict['geocode'] = row[indices['geo_code_index']]
geocode_dict['region'] = row[indices['region_index']]
geocode_dict['city'] = row[indices['city_index']]
object['geocode'] = geocode_dict
return (object, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
geocode_dict = {}
geocode_dict['id'] = row[indices['geo_code_id_index']]
geocode_dict['geocode'] = row[indices['geo_code_index']]
geocode_dict['region'] = row[indices['region_index']]
geocode_dict['city'] = row[indices['city_index']]
object['geocode'] = geocode_dict
return(object, current_id)
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return(object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
reference_id = row[indices['reference_id_index']]
title = row[indices['title_index']]
boiler_text = row[indices['boiler_text_index']]
date = row[indices['date_index']]
main_url = get_media_url(row[indices['path_index']], row[indices['file_index']])
if 'pubdocs' not in object['relateditems']:
object['relateditems']['pubdocs'] = []
object['relateditems']['pubdocs'].append({
'id' : reference_id,
'boilertext' : boiler_text,
'displaytext' : boiler_text,
'date' : date,
'url' : main_url})
# keep the related items sorted
object['relateditems']['pubdocs'].sort(key=operator.itemgetter('displaytext'))
return(object, current_id)
def create_library():
print "Creating Digital Library..."
author_ids = []
size = 20
results_from = 0
es = elasticsearch_connection.get_connection()
es_index = elasticsearch_connection.ELASTICSEARCH_INDEX
# delete library
results = es.search(index=es_index, doc_type='library', body={
"size" : 500,
"fields" : ["_id", "name"],
"query": {
"match_all" : {}
}
})['hits']['hits']
for r in results:
elasticsearch_connection.delete(r['_id'], 'library')
total = es.search(index=es_index, doc_type='pubdocs', body={
"size" : 0,
"query": {
"match_all" : {}
}
})['hits']['total']
while results_from < total:
results = es.search(index=es_index, doc_type='pubdocs', body={
"size" : size,
"from" : results_from,
"query": {
"match_all" : {}
}
})
for r in results['hits']['hits']:
result = r['_source']
if 'pdf' not in result or result['pdf'] == '':
continue
authors = result['authors']
# if this doc has no authors, set the author to 'No Author' and proceed
if len(authors) == 0:
authors.append('No Author')
for author in authors:
author_id = author.replace(' ', '')
# see if this author already exists
if author_id in author_ids:
author_data = elasticsearch_connection.get_item(author_id, 'library')
else:
author_ids.append(author_id)
author_data = {}
author_data['name'] = author
author_data['docs'] = []
author_data['docs'].append({
'displaytext' : result['boilertext'],
'format' : result['format'],
# add file size
'url' : result['pdf']
})
author_data['docs'].sort(key=operator.itemgetter('displaytext'))
data = json.dumps(author_data)
elasticsearch_connection.add_or_update_item(author_id, data, 'library')
results_from = results_from + size
print "Finished Digital Library..."
def process_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
if id != current_id:
# may have multiple rows for one object because of many related constituents
save(object)
current_id = id
object = {}
if elasticsearch_connection.item_exists(id, classification):
object = elasticsearch_connection.get_item(id, classification)
else:
print "%s could not be found!" % id
return (object, current_id)
if 'relateditems' not in object:
object['relateditems'] = {}
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
number = "" if row[indices['rendition_number_index']].lower(
) == "null" else row[indices['rendition_number_index']]
media_master_id = row[indices['media_master_id_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
main_url = get_media_url(row[indices['main_path_index']],
row[indices['main_file_index']])
description = "" if row[indices['description_index']].lower(
) == "null" else row[indices['description_index']]
mediaview = "" if row[indices['media_view_index']].lower(
) == "null" else row[indices['media_view_index']]
caption = "" if row[indices['caption_index']].lower(
) == "null" else row[indices['caption_index']]
display_text = ": ".join([mediaview, caption])
# this is a bit of a hack because the MediaFormats for videos (in the TMS database) does not correctly identify the type of video
# so, make sure we are only using videos that are mp4s
if media_type_key == 3:
if not row[indices['main_file_index']].endswith('mp4'):
return (object, current_id)
if media_type not in object['relateditems']:
object['relateditems'][media_type] = []
if media_type == 'photos':
object['hasphoto'] = "Yes"
# add primary photo as a top level item as well
if row[indices['primary_display_index']] == '1':
object['primarydisplay'] = {
'thumbnail': thumbnail_url,
'main': main_url,
'displaytext': display_text,
'number': number,
'description': description
}
if not (classification == '3dmodels' and media_type == '3dmodels'):
object['relateditems'][media_type].append({
'id':
media_master_id,
'displaytext':
display_text,
'primarydisplay':
True
if row[indices['primary_display_index']] == '1' else False,
'thumbnail':
thumbnail_url,
'main':
main_url,
'number':
number,
'description':
description
})
return (object, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related constituents
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites',
ELASTICSEARCH_INDEX):
site = elasticsearch_connection.get_item(
site_id, 'sites', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % site_id)
return (site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
constituent_id = row[indices['constituent_id_index']]
display_name = row[indices['display_name_index']]
display_date = ""
if row[indices['display_date_index']] != "NULL":
display_date = row[indices['display_date_index']]
drs_id = "" if row[indices['drs_id']].lower() == "null" else row[
indices['drs_id']]
has_manifest = False if drs_id == "" else True
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
if not thumbnail_url and drs_id:
thumbnail_url = create_thumbnail_url(drs_id)
constituent_dict = {}
role = row[indices['role_index']]
# update the set of roles for this site
if role not in site['roles']:
# make sure Tomb Owner is first
if role == "Tomb Owner":
site['roles'].insert(0, role)
else:
site['roles'].append(role)
description = row[indices['remarks_index']] if row[
indices['remarks_index']] != "NULL" else ""
constituent_dict['role'] = role
constituent_dict['id'] = constituent_id
constituent_dict['displayname'] = display_name
constituent_dict['displaydate'] = display_date
constituent_dict['displaytext'] = display_name
constituent_dict['description'] = description
constituent_dict['thumbnail'] = thumbnail_url
constituent_dict['has_manifest'] = has_manifest
constituent_type_key = int(row[indices['constituent_type_id_index']])
constituent_type = CONSTITUENTTYPES.get(constituent_type_key)
# add to array of people for easier searching
if (constituent_type_key in [1, 3]):
site['people'].append(display_name)
if constituent_type not in site['relateditems']:
site['relateditems'][constituent_type] = []
site['relateditems'][constituent_type].append(constituent_dict)
# keep the related items sorted
site['relateditems'][constituent_type].sort(
key=operator.itemgetter('displaytext'))
if role == 'Tomb Owner':
site['tombowner'] = "Yes"
return (site, current_id)
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related photos
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites',
ELASTICSEARCH_INDEX):
site = elasticsearch_connection.get_item(
site_id, 'sites', ELASTICSEARCH_INDEX)
else:
print("%s could not be found!" % site_id)
return (site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
number = "" if row[indices['rendition_number_index']].lower(
) == "null" else row[indices['rendition_number_index']]
description = "" if row[indices['description_index']].lower(
) == "null" else row[indices['description_index']]
mediaview = "" if row[indices['media_view_index']].lower(
) == "null" else row[indices['media_view_index']]
caption = "" if row[indices['caption_index']].lower(
) == "null" else row[indices['caption_index']]
display_text = ": ".join([mediaview, caption])
media_master_id = row[indices['media_master_id_index']]
main_url = get_media_url(row[indices['main_path_index']],
row[indices['main_file_index']])
drs_id = "" if row[indices['drs_id']].lower() == "null" else row[
indices['drs_id']]
has_manifest = False if drs_id == "" else True
primary_display = True if row[
indices['primary_display_index']] == '1' else False
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
if not thumbnail_url and drs_id:
thumbnail_url = create_thumbnail_url(drs_id)
# this is a bit of a hack because the MediaFormats for videos (in the TMS database) does not correctly identify the type of video
# so, make sure we are only using videos that are mp4s
if media_type_key == 3:
if not row[indices['main_file_index']].endswith('mp4'):
return (site, current_id)
if media_type not in site['relateditems']:
site['relateditems'][media_type] = []
# add primary photo as a top level item as well
if primary_display:
site['primarydisplay'] = {
'thumbnail': thumbnail_url,
'main': main_url,
'displaytext': display_text,
'number': number,
'description': description,
'has_manifest': has_manifest,
'media_id': media_master_id
}
site['relateditems'][media_type].append({
'id': media_master_id,
'displaytext': display_text,
'primarydisplay': primary_display,
'thumbnail': thumbnail_url,
'main': main_url,
'number': number,
'description': description,
'has_manifest': has_manifest,
'drs_id': drs_id
})
if has_manifest:
object = elasticsearch_connection.get_item(
media_type + '-' + media_master_id, 'manifest',
ELASTICSEARCH_IIIF_INDEX)
resource = object['manifest']['sequences'][0]['canvases'][0][
'images'][0]['resource']
canvas_label = object['manifest']['description']
canvas_metadata = object['manifest'][
'metadata'] #add photo manifest-level metadata as canvas-level metadata for site
if site_id not in SITE_RELATIONS.keys():
metadata = add_metadata_to_manifest(site)
SITE_RELATIONS[site_id] = {
'description': site['description'],
'label': site['displaytext'],
'resources': [resource],
'drs_ids': [drs_id],
'canvas_labels': [canvas_label],
'canvas_metadatas': [canvas_metadata],
'metadata': metadata
}
else:
SITE_RELATIONS[site_id]['resources'].append(resource)
SITE_RELATIONS[site_id]['drs_ids'].append(drs_id)
SITE_RELATIONS[site_id]['canvas_labels'].append(canvas_label)
SITE_RELATIONS[site_id]['canvas_metadatas'].append(
canvas_metadata)
if primary_display:
SITE_RELATIONS[site_id]['startCanvas'] = drs_id
return (site, current_id)
def create_library():
print("Creating Digital Library...")
time.sleep(
3
) # for some reason the library isn't always fully populated. see if a time delay helps
author_ids = []
size = 20
results_from = 0
es = elasticsearch_connection.get_connection()
es_index = ELASTICSEARCH_INDEX
# delete library
results = es.search(index=es_index,
doc_type='library',
body={
"size": 500,
"stored_fields": ["_id", "name"],
"query": {
"match_all": {}
}
})['hits']['hits']
for r in results:
elasticsearch_connection.delete(r['_id'], 'library',
ELASTICSEARCH_INDEX)
total = es.search(index=es_index,
doc_type='pubdocs',
body={
"size": 0,
"query": {
"match_all": {}
}
})['hits']['total']
while results_from < total:
results = es.search(index=es_index,
doc_type='pubdocs',
body={
"size": size,
"from": results_from,
"query": {
"match_all": {}
}
})
for r in results['hits']['hits']:
result = r['_source']
if 'pdf' not in result or result['pdf'] == '':
continue
authors = result['authors']
# if this doc has no authors, set the author to 'No Author' and proceed
if len(authors) == 0:
authors.append('No Author')
for author in authors:
author_id = author.replace(' ', '')
sortauthor = author.lower().strip()
sortauthor = str(
unicodedata.normalize('NFD', sortauthor).encode(
'ascii', 'ignore').decode("utf-8"))
# see if this author already exists
if author_id in author_ids:
author_data = elasticsearch_connection.get_item(
author_id, 'library', ELASTICSEARCH_INDEX)
else:
author_ids.append(author_id)
author_data = {}
author_data['name'] = author
author_data['sortname'] = sortauthor
author_data['docs'] = []
author_data['docs'].append({
'displaytext':
result['boilertext'],
'sorttext':
result['notes']
if result['notes'] is not None else result['title'],
'format':
result['format'],
# add file size
'url':
result['pdf']
})
author_data['docs'].sort(key=operator.itemgetter('sorttext'))
data = json.dumps(author_data)
elasticsearch_connection.add_or_update_item(
author_id, data, 'library', ELASTICSEARCH_INDEX)
results_from = results_from + size
print("Finished Digital Library...")
def process_site_row(site, current_id):
site_id = row[indices['site_id_index']]
#if site_id not in SAMPLE_SITES:
# continue
if site_id != current_id:
# will likely have multiple rows for one site because of many related photos
# only get a new site if we have a new site id, but first save old site to elasticsearch
save(site)
current_id = site_id
site = {}
if elasticsearch_connection.item_exists(site_id, 'sites'):
site = elasticsearch_connection.get_item(site_id, 'sites')
else:
print "%s could not be found!" % site_id
return (site, current_id)
if 'relateditems' not in site:
site['relateditems'] = {}
media_type_key = int(row[indices['media_type_id_index']])
media_type = MEDIATYPES.get(media_type_key)
number = "" if row[indices['rendition_number_index']].lower(
) == "null" else row[indices['rendition_number_index']]
description = "" if row[indices['description_index']].lower(
) == "null" else row[indices['description_index']]
mediaview = "" if row[indices['media_view_index']].lower(
) == "null" else row[indices['media_view_index']]
caption = "" if row[indices['caption_index']].lower(
) == "null" else row[indices['caption_index']]
display_text = ": ".join([mediaview, caption])
media_master_id = row[indices['media_master_id_index']]
thumbnail_url = get_media_url(row[indices['thumb_path_index']],
row[indices['thumb_file_index']])
main_url = get_media_url(row[indices['main_path_index']],
row[indices['main_file_index']])
# this is a bit of a hack because the MediaFormats for videos (in the TMS database) does not correctly identify the type of video
# so, make sure we are only using videos that are mp4s
if media_type_key == 3:
if not row[indices['main_file_index']].endswith('mp4'):
return (site, current_id)
if media_type not in site['relateditems']:
site['relateditems'][media_type] = []
# add primary photo as a top level item as well
if row[indices['primary_display_index']] == '1':
site['primarydisplay'] = {
'thumbnail': thumbnail_url,
'main': main_url,
'displaytext': display_text,
'number': number,
'description': description
}
site['relateditems'][media_type].append({
'id':
media_master_id,
'displaytext':
display_text,
'primarydisplay':
True if row[indices['primary_display_index']] == '1' else False,
'thumbnail':
thumbnail_url,
'main':
main_url,
'number':
number,
'description':
description
})
return (site, current_id)
