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_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 '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_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:
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_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'] = {}
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_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_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_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_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_object_row(object, current_id):
id = row[indices['id_index']]
classification_key = int(row[indices['classification_id_index']])
classification = CLASSIFICATIONS.get(classification_key)
number = row[indices['object_number_index']]
#if id not in SAMPLE_OBJECTS:
# return (object, current_id)
# I don't think there are any duplicate rows for objects, but keep it here since it doesn't break anything
if id != current_id:
save(object)
current_id = id
object = {}
object['classification'] = classification
# loop through each row
for index, value in enumerate(columns):
key = value.lower()
row_value = row[index]
# cleanup row data
if row_value.isdigit():
row_value = int(row_value)
elif row_value == "NULL":
row_value = None
else:
row_value = row_value.replace(',,', '')
if 'title' in key:
object_title = row_value
if classification == "diarypages" and object_title is None:
idx = number.find('_')
object_title = number[idx + 1:]
object[key] = object_title
else:
object[key] = row_value
else:
# remove whitespace at end of line if a string
object[key] = row_value.rstrip(
) if type(row_value) is str else row_value
# Add some extra fields not in the TMS data
object['displaytext'] = object['title']
prefix_idx = number.find('_')
object['allnumbers'] = list(
set([number, number[prefix_idx + 1:], "".join(number.split())]))
object['roles'] = []
object['hasphoto'] = "No"
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)
object_number = row[indices['object_number_index']]
#if id not in SAMPLE_OBJECTS:
# return (object, current_id)
# I don't think there are any duplicate rows for objects, but keep it here since it doesn't break anything
if id != current_id:
save(object)
current_id = id
object = {}
object['classification'] = classification
# loop through each row
for index, value in enumerate(columns):
key = value.lower()
row_value = row[index]
# cleanup row data
if row_value.isdigit():
row_value = int(row_value)
elif row_value == "NULL":
row_value = None
else:
row_value = row_value.replace(',,','')
if 'title' in key:
object_title = row_value
if classification == "diarypages" and object_title is None:
idx = object_number.find('_')
object_title = object_number[idx+1:]
object[key] = object_title
else:
object[key] = row_value
else:
# remove whitespace at end of line if a string
object[key] = row_value.rstrip() if type(row_value) is str else row_value
# Add some extra fields not in the TMS data
object['displaytext'] = object['title']
prefix_idx = object_number.find('_')
object['allnumbers'] = [object_number, object_number[prefix_idx+1:]]
object['roles'] = []
object['hasphoto'] = "No"
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'] = {}
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_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):
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_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_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_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_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_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 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)
