def find_series_for_mail(project, name, msgid, is_patch, order, refs):
if refs == []:
root_msgid = msgid
else:
root_msgid = refs[-1]
try:
# grab the latest revision for this mail thread
revisions = SeriesRevision.objects.filter(series__project=project,
root_msgid=root_msgid) \
.reverse()
revision = revisions[0]
series = revision.series
if name:
series.name = name
if is_patch:
previous_patch = find_previous_patch(revision, order, refs)
if previous_patch:
order = find_patch_order(revisions, previous_patch, order)
revision = revision.duplicate(exclude_patches=(order, ))
# series has been updated, grab the new instance
series = revision.series
except IndexError:
if not name:
name = SERIES_DEFAULT_NAME
series = Series(name=name)
revision = SeriesRevision(root_msgid=root_msgid)
return (series, revision, order)
def find_series_for_mail(project, name, msgid, is_patch, order, n_patches,
refs, is_cover_letter):
if refs == []:
root_msgid = msgid
else:
root_msgid = refs[-1]
try:
# grab the latest revision for this mail thread
revisions = SeriesRevision.objects.filter(series__project=project,
root_msgid=root_msgid) \
.reverse()
revision = revisions[0]
series = revision.series
if name:
series.name = name
if is_patch:
previous_patch = find_previous_patch(project, revision, order,
refs)
if previous_patch:
(order, n_patches) = find_patch_order(revisions,
previous_patch, order,
n_patches)
revision = revision.duplicate(exclude_patches=(order, ))
# series has been updated, grab the new instance
series = revision.series
else:
try:
prev_patch = SeriesRevisionPatch.objects.get(
revision=revision, order=order).patch
if prev_patch and not prev_patch.msgid == msgid:
# this is a new patch in the thread
revision = revision.duplicate(
exclude_patches=(order, ))
series = revision.series
except SeriesRevisionPatch.DoesNotExist:
pass
elif is_cover_letter and not refs == []:
revision = revision.duplicate(exclude_patches=range(
1, revision.n_patches + 1), )
series = revision.series
except IndexError:
if not name:
name = SERIES_DEFAULT_NAME
series = Series(name=name)
revision = SeriesRevision(root_msgid=root_msgid)
return (series, revision, order, n_patches)
def parse_mail(mail, list_id=None):
"""Parse a mail and add to the database.
Args:
mail (`mbox.Mail`): Mail to parse and add.
list_id (str): Mailing list ID
Returns:
None
"""
# some basic sanity checks
if 'From' not in mail:
raise ValueError("Missing 'From' header")
if 'Subject' not in mail:
raise ValueError("Missing 'Subject' header")
if 'Message-Id' not in mail:
raise ValueError("Missing 'Message-Id' header")
hint = clean_header(mail.get('X-Patchwork-Hint', ''))
if hint and hint.lower() == 'ignore':
logger.debug("Ignoring email due to 'ignore' hint")
return
if list_id:
project = find_project_by_id(list_id)
else:
project = find_project_by_header(mail)
if project is None:
logger.error('Failed to find a project for email')
return
# parse metadata
msgid = clean_header(mail.get('Message-Id'))
if not msgid:
raise ValueError("Broken 'Message-Id' header")
msgid = msgid[:255]
author = find_author(mail)
subject = mail.get('Subject')
name, prefixes = clean_subject(subject, [project.linkname])
is_comment = subject_check(subject)
x, n = parse_series_marker(prefixes)
version = parse_version(name, prefixes)
refs = find_references(mail)
date = find_date(mail)
headers = find_headers(mail)
# parse content
if not is_comment:
diff, message = find_patch_content(mail)
else:
diff, message = find_comment_content(mail)
if not (diff or message):
return # nothing to work with
pull_url = parse_pull_request(message)
# build objects
if not is_comment and (diff or pull_url): # patches or pull requests
# we delay the saving until we know we have a patch.
author.save()
delegate = find_delegate_by_header(mail)
if not delegate and diff:
filenames = find_filenames(diff)
delegate = find_delegate_by_filename(project, filenames)
# if we don't have a series marker, we will never have an existing
# series to match against.
series = None
if n:
series = find_series(project, mail)
else:
x = n = 1
# We will create a new series if:
# - there is no existing series to assign this patch to, or
# - there is an existing series, but it already has a patch with this
# number in it
if not series or (
SeriesPatch.objects.filter(series=series, number=x).count()):
series = Series(project=project,
date=date,
submitter=author,
version=version,
total=n)
series.save()
# NOTE(stephenfin) We must save references for series. We
# do this to handle the case where a later patch is
# received first. Without storing references, it would not
# be possible to identify the relationship between patches
# as the earlier patch does not reference the later one.
for ref in refs + [msgid]:
ref = ref[:255]
# we don't want duplicates
try:
# we could have a ref to a previous series. (For
# example, a series sent in reply to another
# series.) That should not create a series ref
# for this series, so check for the msg-id only,
# not the msg-id/series pair.
SeriesReference.objects.get(msgid=ref,
series__project=project)
except SeriesReference.DoesNotExist:
SeriesReference.objects.create(series=series, msgid=ref)
patch = Patch(
msgid=msgid,
project=project,
name=name[:255],
date=date,
headers=headers,
submitter=author,
content=message,
diff=diff,
pull_url=pull_url,
delegate=delegate,
state=find_state(mail))
patch.save()
logger.debug('Patch saved')
# add to a series if we have found one, and we have a numbered
# patch. Don't add unnumbered patches (for example diffs sent
# in reply, or just messages with random refs/in-reply-tos)
if series and x:
series.add_patch(patch, x)
return patch
elif x == 0: # (potential) cover letters
# if refs are empty, it's implicitly a cover letter. If not,
# however, we need to see if a match already exists and, if
# not, assume that it is indeed a new cover letter
is_cover_letter = False
if not is_comment:
if not refs == []:
try:
CoverLetter.objects.all().get(name=name)
except CoverLetter.DoesNotExist:
# if no match, this is a new cover letter
is_cover_letter = True
except CoverLetter.MultipleObjectsReturned:
# if multiple cover letters are found, just ignore
pass
else:
is_cover_letter = True
if is_cover_letter:
author.save()
# we don't use 'find_series' here as a cover letter will
# always be the first item in a thread, thus the references
# could only point to a different series or unrelated
# message
try:
series = SeriesReference.objects.get(
msgid=msgid, series__project=project).series
except SeriesReference.DoesNotExist:
series = None
if not series:
series = Series(project=project,
date=date,
submitter=author,
version=version,
total=n)
series.save()
# we don't save the in-reply-to or references fields
# for a cover letter, as they can't refer to the same
# series
SeriesReference.objects.get_or_create(series=series,
msgid=msgid)
cover_letter = CoverLetter(
msgid=msgid,
project=project,
name=name[:255],
date=date,
headers=headers,
submitter=author,
content=message)
cover_letter.save()
logger.debug('Cover letter saved')
series.add_cover_letter(cover_letter)
return cover_letter
# comments
# we only save comments if we have the parent email
submission = find_submission_for_comment(project, refs)
if not submission:
return
author.save()
comment = Comment(
submission=submission,
msgid=msgid,
date=date,
headers=headers,
submitter=author,
content=message)
comment.save()
logger.debug('Comment saved')
return comment
def parse_mail(mail, list_id=None):
"""Parse a mail and add to the database.
Args:
mail (`mbox.Mail`): Mail to parse and add.
list_id (str): Mailing list ID
Returns:
patch/cover letter/comment
Or None if nothing is found in the mail
or X-P-H: ignore
or project not found
Raises:
ValueError if there is an error in parsing or a duplicate mail
Other truly unexpected issues may bubble up from the DB.
"""
# some basic sanity checks
if 'From' not in mail:
raise ValueError("Missing 'From' header")
if 'Subject' not in mail:
raise ValueError("Missing 'Subject' header")
if 'Message-Id' not in mail:
raise ValueError("Missing 'Message-Id' header")
hint = clean_header(mail.get('X-Patchwork-Hint', ''))
if hint and hint.lower() == 'ignore':
logger.debug("Ignoring email due to 'ignore' hint")
return
project = find_project(mail, list_id)
if project is None:
logger.error('Failed to find a project for email')
return
# parse metadata
msgid = clean_header(mail.get('Message-Id'))
if not msgid:
raise ValueError("Broken 'Message-Id' header")
msgid = msgid[:255]
subject = mail.get('Subject')
name, prefixes = clean_subject(subject, [project.linkname])
is_comment = subject_check(subject)
x, n = parse_series_marker(prefixes)
version = parse_version(name, prefixes)
refs = find_references(mail)
date = find_date(mail)
headers = find_headers(mail)
# parse content
if not is_comment:
diff, message = find_patch_content(mail)
else:
diff, message = find_comment_content(mail)
if not (diff or message):
return # nothing to work with
pull_url = parse_pull_request(message)
# build objects
if not is_comment and (diff or pull_url): # patches or pull requests
# we delay the saving until we know we have a patch.
author = get_or_create_author(mail)
delegate = find_delegate_by_header(mail)
if not delegate and diff:
filenames = find_filenames(diff)
delegate = find_delegate_by_filename(project, filenames)
try:
patch = Patch.objects.create(msgid=msgid,
project=project,
patch_project=project,
name=name[:255],
date=date,
headers=headers,
submitter=author,
content=message,
diff=diff,
pull_url=pull_url,
delegate=delegate,
state=find_state(mail))
logger.debug('Patch saved')
except IntegrityError:
logger.error("Duplicate mail for message ID %s" % msgid)
return None
# if we don't have a series marker, we will never have an existing
# series to match against.
series = None
if n:
series = find_series(project, mail, author)
else:
x = n = 1
# We will create a new series if:
# - there is no existing series to assign this patch to, or
# - there is an existing series, but it already has a patch with this
# number in it
if not series or (SeriesPatch.objects.filter(series=series,
number=x).count()):
series = Series(project=project,
date=date,
submitter=author,
version=version,
total=n)
series.save()
# NOTE(stephenfin) We must save references for series. We
# do this to handle the case where a later patch is
# received first. Without storing references, it would not
# be possible to identify the relationship between patches
# as the earlier patch does not reference the later one.
for ref in refs + [msgid]:
ref = ref[:255]
# we don't want duplicates
try:
# we could have a ref to a previous series. (For
# example, a series sent in reply to another
# series.) That should not create a series ref
# for this series, so check for the msg-id only,
# not the msg-id/series pair.
SeriesReference.objects.get(msgid=ref,
series__project=project)
except SeriesReference.DoesNotExist:
SeriesReference.objects.create(series=series, msgid=ref)
except SeriesReference.MultipleObjectsReturned:
logger.error("Multiple SeriesReferences for %s"
" in project %s!" % (ref, project.name))
# add to a series if we have found one, and we have a numbered
# patch. Don't add unnumbered patches (for example diffs sent
# in reply, or just messages with random refs/in-reply-tos)
if series and x:
series.add_patch(patch, x)
return patch
elif x == 0: # (potential) cover letters
# if refs are empty, it's implicitly a cover letter. If not,
# however, we need to see if a match already exists and, if
# not, assume that it is indeed a new cover letter
is_cover_letter = False
if not is_comment:
if not refs == []:
try:
CoverLetter.objects.all().get(name=name)
except CoverLetter.DoesNotExist:
# if no match, this is a new cover letter
is_cover_letter = True
except CoverLetter.MultipleObjectsReturned:
# if multiple cover letters are found, just ignore
pass
else:
is_cover_letter = True
if is_cover_letter:
author = get_or_create_author(mail)
# we don't use 'find_series' here as a cover letter will
# always be the first item in a thread, thus the references
# could only point to a different series or unrelated
# message
try:
series = SeriesReference.objects.get(
msgid=msgid, series__project=project).series
except SeriesReference.DoesNotExist:
series = None
except SeriesReference.MultipleObjectsReturned:
logger.error("Multiple SeriesReferences for %s"
" in project %s!" % (msgid, project.name))
series = SeriesReference.objects.filter(
msgid=msgid, series__project=project).first().series
if not series:
series = Series(project=project,
date=date,
submitter=author,
version=version,
total=n)
series.save()
# we don't save the in-reply-to or references fields
# for a cover letter, as they can't refer to the same
# series
try:
SeriesReference.objects.get_or_create(series=series,
msgid=msgid)
except SeriesReference.MultipleObjectsReturned:
logger.error("Multiple SeriesReferences for %s"
" in project %s!" % (msgid, project.name))
cover_letter = CoverLetter(msgid=msgid,
project=project,
name=name[:255],
date=date,
headers=headers,
submitter=author,
content=message)
cover_letter.save()
logger.debug('Cover letter saved')
series.add_cover_letter(cover_letter)
return cover_letter
# comments
# we only save comments if we have the parent email
submission = find_submission_for_comment(project, refs)
if not submission:
return
author = get_or_create_author(mail)
comment = Comment(submission=submission,
msgid=msgid,
date=date,
headers=headers,
submitter=author,
content=message)
comment.save()
logger.debug('Comment saved')
return comment
def parse_mail(mail, list_id=None):
"""Parse a mail and add to the database.
Args:
mail (`mbox.Mail`): Mail to parse and add.
list_id (str): Mailing list ID
Returns:
None
"""
# some basic sanity checks
if 'From' not in mail:
raise ValueError("Missing 'From' header")
if 'Subject' not in mail:
raise ValueError("Missing 'Subject' header")
if 'Message-Id' not in mail:
raise ValueError("Missing 'Message-Id' header")
hint = mail.get('X-Patchwork-Hint', '').lower()
if hint == 'ignore':
logger.debug("Ignoring email due to 'ignore' hint")
return
if list_id:
project = find_project_by_id(list_id)
else:
project = find_project_by_header(mail)
if project is None:
logger.error('Failed to find a project for email')
return
# parse content
diff, message = find_content(project, mail)
if not (diff or message):
return # nothing to work with
msgid = mail.get('Message-Id').strip()
author = find_author(mail)
subject = mail.get('Subject')
name, prefixes = clean_subject(subject, [project.linkname])
is_comment = subject_check(subject)
x, n = parse_series_marker(prefixes)
version = parse_version(name, prefixes)
refs = find_references(mail)
date = find_date(mail)
headers = find_headers(mail)
pull_url = parse_pull_request(message)
# build objects
if not is_comment and (diff or pull_url): # patches or pull requests
# we delay the saving until we know we have a patch.
author.save()
delegate = find_delegate(mail)
if not delegate and diff:
filenames = find_filenames(diff)
delegate = auto_delegate(project, filenames)
series = find_series(mail)
# We will create a new series if:
# - we have a patch number (x of n), and
# - either:
# * there is no series, or
# * the version doesn't match
# * we have a patch with this number already
if n and (
(not series) or (series.version != version) or
(SeriesPatch.objects.filter(series=series, number=x).count())):
series = Series(date=date,
submitter=author,
version=version,
total=n)
series.save()
# NOTE(stephenfin) We must save references for series. We
# do this to handle the case where a later patch is
# received first. Without storing references, it would not
# be possible to identify the relationship between patches
# as the earlier patch does not reference the later one.
for ref in refs + [msgid]:
# we don't want duplicates
try:
# we could have a ref to a previous series. (For
# example, a series sent in reply to another
# series.) That should not create a series ref
# for this series, so check for the msg-id only,
# not the msg-id/series pair.
SeriesReference.objects.get(msgid=ref)
except SeriesReference.DoesNotExist:
SeriesReference.objects.create(series=series, msgid=ref)
patch = Patch(msgid=msgid,
project=project,
name=name,
date=date,
headers=headers,
submitter=author,
content=message,
diff=diff,
pull_url=pull_url,
delegate=delegate,
state=find_state(mail))
patch.save()
logger.debug('Patch saved')
# add to a series if we have found one, and we have a numbered
# patch. Don't add unnumbered patches (for example diffs sent
# in reply, or just messages with random refs/in-reply-tos)
if series and x:
series.add_patch(patch, x)
return patch
elif x == 0: # (potential) cover letters
# if refs are empty, it's implicitly a cover letter. If not,
# however, we need to see if a match already exists and, if
# not, assume that it is indeed a new cover letter
is_cover_letter = False
if not is_comment:
if not refs == []:
try:
CoverLetter.objects.all().get(name=name)
except CoverLetter.DoesNotExist:
# if no match, this is a new cover letter
is_cover_letter = True
except CoverLetter.MultipleObjectsReturned:
# if multiple cover letters are found, just ignore
pass
else:
is_cover_letter = True
if is_cover_letter:
author.save()
# we don't use 'find_series' here as a cover letter will
# always be the first item in a thread, thus the references
# could only point to a different series or unrelated
# message
try:
series = SeriesReference.objects.get(msgid=msgid).series
except SeriesReference.DoesNotExist:
series = None
if not series:
series = Series(date=date,
submitter=author,
version=version,
total=n)
series.save()
# we don't save the in-reply-to or references fields
# for a cover letter, as they can't refer to the same
# series
SeriesReference.objects.get_or_create(series=series,
msgid=msgid)
cover_letter = CoverLetter(msgid=msgid,
project=project,
name=name,
date=date,
headers=headers,
submitter=author,
content=message)
cover_letter.save()
logger.debug('Cover letter saved')
series.add_cover_letter(cover_letter)
return cover_letter
# comments
# we only save comments if we have the parent email
submission = find_submission_for_comment(project, refs)
if not submission:
return
if is_comment and diff:
message += diff
author.save()
comment = Comment(submission=submission,
msgid=msgid,
date=date,
headers=headers,
submitter=author,
content=message)
comment.save()
logger.debug('Comment saved')
return comment