tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Hanword: use the html converter, the text ones drops data from tables

Browse Source
This commit is contained in:
Jean-Francois Dockes 2020-03-21 10:16:16 +01:00
parent d2b695d705
commit 03cbc203e1
2 changed files with 9 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
src/filters/rclhwp.py
View File

@ -25,6 +25,7 @@
import sys import sys
from io import BytesIO from io import BytesIO
import subprocess
import rclexecm import rclexecm
from rclbasehandler import RclBaseHandler from rclbasehandler import RclBaseHandler
@ -68,16 +69,12 @@ class HWP5Dump(RclBaseHandler):
# hwp wants str filenames. This is unfortunate # hwp wants str filenames. This is unfortunate
fn = fn.decode('utf-8') fn = fn.decode('utf-8')
html = b'<html><head>\n' + \
b'<meta http-equiv="content-type" \
content="text/html; charset=utf-8">\n'
hwpfile = fs_Hwp5File(fn) hwpfile = fs_Hwp5File(fn)
try: try:
tt = hwpfile.summaryinfo.title.strip() tt = hwpfile.summaryinfo.title.strip()
if tt: if tt:
tt = self.em.htmlescape(tt.encode('utf-8')) tt = self.em.htmlescape(tt.encode('utf-8'))
html += b'<title>' + tt + b'</title>\n' self.em.setfield('caption', tt)
for k,v in metafields(hwpfile.summaryinfo): for k,v in metafields(hwpfile.summaryinfo):
v = "{0}".format(v) v = "{0}".format(v)
@ -85,28 +82,17 @@ class HWP5Dump(RclBaseHandler):
if v: if v:
v = self.em.htmlescape(v.encode('utf-8')) v = self.em.htmlescape(v.encode('utf-8'))
k = k.encode('utf-8') k = k.encode('utf-8')
html += b'<meta name="' + k + b'" content="' + \ self.em.setfield(k, v)
v + b'">\n'
except Exception as e: except Exception as e:
self.em.rclog("Exception: %s" % e) self.em.rclog("Exception: %s" % e)
finally: finally:
hwpfile.close() hwpfile.close()
html += b'</head><body><pre>\n' # The first version of this file used conversion to text using
# the hwp5 module (no subproc). But this apparently mishandled
hwpfile = xml_Hwp5File(fn) # tables. Switched to executing hwp5html instead. See 1st git
text_transform = TextTransform() # version for the old approach.
transform = text_transform.transform_hwp5_to_text html = subprocess.check_output(["hwp5html", "--html", fn])
dest = BytesIO()
try:
transform(hwpfile, dest)
except Exception as e:
self.em.rclog("Exception: %s" % e)
finally:
hwpfile.close()
dest.seek(0)
html += self.em.htmlescape(dest.read())
html += b'</pre></body></html>'
return html return html
if __name__ == '__main__': if __name__ == '__main__':

2
src/sampleconf/mimeview
View File

@ -43,7 +43,7 @@ application/x-mobipocket-ebook = ebook-viewer %f
application/x-kword = kword %f application/x-kword = kword %f
application/x-abiword = abiword %f application/x-abiword = abiword %f
application/x-hwp = evince-hwp application/x-hwp = libreoffice
# Note: the Linux Mint evince clones, atril and xread, have the same options # Note: the Linux Mint evince clones, atril and xread, have the same options
application/pdf = evince --page-index=%p --find=%s %f application/pdf = evince --page-index=%p --find=%s %f