Show unmentioned images in issue description (#135)
* show unmentioned images in issue description
* support other image file types
diff --git a/migration/src/jira2github_import.py b/migration/src/jira2github_import.py
index 18d0ad2..5ba69c0 100644
--- a/migration/src/jira2github_import.py
+++ b/migration/src/jira2github_import.py
@@ -61,6 +61,7 @@
priority = extract_priority(o)
vote_count = extract_vote_count(o)
parent_issue_key = extract_parent_key(o)
+ comments = extract_comments(o)
reporter_gh = account_map.get(reporter_name)
reporter = f"{reporter_dispname} (@{reporter_gh})" if reporter_gh else f"{reporter_dispname}"
@@ -74,6 +75,14 @@
attachment_list_items.append(f"[{filename}]({attachment_url(num, filename, att_repo, att_branch)})" + (f" (versions: {cnt})" if cnt > 1 else ""))
att_replace_map[filename] = attachment_url(num, filename, att_repo, att_branch)
+ # detect unmentioned image files
+ # https://github.com/apache/lucene-jira-archive/issues/126
+ image_files = [x[0] for x in attachments if re.match(r"^.+\.(png|jpg|jpeg|gif|svg|bmp|ico|tif|tiff)$", x[0], flags=re.IGNORECASE)]
+ embedded_image_files = extract_embedded_image_files(description, image_files)
+ for (_, _, comment_body, _, _, _) in comments:
+ embedded_image_files.update(extract_embedded_image_files(comment_body, image_files))
+ unmentioned_images = [x for x in image_files if x not in embedded_image_files]
+
# embed github issue number next to linked issue keys
linked_issues_list_items = []
for jira_key in linked_issues:
@@ -93,6 +102,10 @@
try:
body = f'{convert_text(description, att_replace_map, account_map, jira_users)}\n\n'
+ for image_file in unmentioned_images:
+ # show orphaned (unmentioned) image files in the issue description
+ att_url = att_replace_map.get(image_file)
+ body += f'\n\n'
except Exception as e:
logger.error(traceback.format_exc(limit=100))
logger.error(f"Failed to convert opening issue description on {jira_issue_id(num)} due to above exception, ({str(e)}); falling back to original Jira description as code block.")
@@ -149,7 +162,6 @@
lines.append(line)
return "\n".join(lines)
- comments = extract_comments(o)
comments_data = []
for (comment_author_name, comment_author_dispname, comment_body, comment_created, comment_updated, comment_id) in comments:
# TODO: since we now have accurate created_at reflected in the github comment, mabye we remove these
diff --git a/migration/src/jira_util.py b/migration/src/jira_util.py
index deb36e7..cf76c77 100644
--- a/migration/src/jira_util.py
+++ b/migration/src/jira_util.py
@@ -205,6 +205,8 @@
return res
+REGEX_EMBEDDED_IMAGE = r"!([^!\n]+)!"
+
# space character + Jira emoji + space character
JIRA_EMOJI_TO_UNICODE = {
"(?<=\s)\(y\)((?=$)|(?=\s))": "\U0001F44D",
@@ -234,6 +236,23 @@
REGEX_GITHUB_ISSUE_LINK = re.compile(r"(\s)(#\d+)(\s)")
+def extract_embedded_image_files(text: str, image_files: list[str]) -> set[str]:
+ """Extract embedded image files in the given text.
+ https://jira.atlassian.com/secure/WikiRendererHelpAction.jspa?section=images
+ """
+ # capture candidates for embedded images
+ candidates = re.findall(REGEX_EMBEDDED_IMAGE, text)
+ embedded_image_files = set([])
+ for x in candidates:
+ if x in image_files:
+ # !xyz.png!
+ embedded_image_files.add(x)
+ elif any(map(lambda s: x.startswith(s + "|"), image_files)):
+ # !xyz.png|styles!
+ embedded_image_files.add(x.split("|", 1)[0])
+ return embedded_image_files
+
+
def convert_text(text: str, att_replace_map: dict[str, str] = {}, account_map: dict[str, str] = {}, jira_users: dict[str, str] = {}) -> str:
"""Convert Jira markup to Markdown
"""
