moparisthebest/xeps
1
0
Fork 0
mirror of https://github.com/moparisthebest/xeps synced 2024-11-25 02:32:18 -05:00
Code Issues Releases Wiki Activity

Support for multi-<p/> revision block remarks

Browse Source
This commit is contained in:
Jonas Wielicki 2017-11-29 11:55:26 +01:00
parent 0fed18bac5
commit 14c40d8620
1 changed files with 17 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
tools/extract-metadata.py
View File

@ -23,6 +23,22 @@ def open_xml(f):
return xml.dom.minidom.parse(f) return xml.dom.minidom.parse(f)
def extract_revision_text(remark_el):
remark_children = minidom_children(remark_el)
if len(remark_children) == 1 and remark_children[0].tagName == "p":
return minidom_get_text(remark_children[0])
if len(remark_children) == 0:
return minidom_get_text(remark_el) or None
if minidom_get_text(remark_el).strip():
return None
lines = []
for child in remark_children:
if child.tagName == "p":
lines.append(minidom_get_text(child))
return "\n".join(lines)
def extract_xep_metadata(document): def extract_xep_metadata(document):
header = minidom_find_header(document) header = minidom_find_header(document)
@ -37,11 +53,7 @@ def extract_xep_metadata(document):
remark_el = minidom_find_child(latest_revision, "remark") remark_el = minidom_find_child(latest_revision, "remark")
last_revision_remark = None last_revision_remark = None
if remark_el is not None: if remark_el is not None:
remark_children = minidom_children(remark_el) last_revision_remark = extract_revision_text(remark_el)
if len(remark_children) == 1 and remark_children[0].tagName == "p":
last_revision_remark = minidom_get_text(remark_children[0])
elif len(remark_children) == 0:
last_revision_remark = minidom_get_text(remark_el) or None
if last_revision_remark is not None: if last_revision_remark is not None:
initials_el = minidom_find_child(latest_revision, "initials") initials_el = minidom_find_child(latest_revision, "initials")