search: Replace third-party OpenDocument search helper.

The odt2txt package is extremely slow when working with larger
documents, and has not seen an update in 7+ years.

Fixes #3481

Author: mtwebster

debian/control

@@ -94,7 +94,6 @@ Depends:
  poppler-utils,
  exif,
  id3,
- odt2txt,
  catdoc,
  untex,
  html2text,

search-helpers/meson.build

@@ -21,7 +21,7 @@ ppt_to_txt = executable('nemo-ppt-to-txt',
 )
 
 install_data(
-    'nemo-xls-to-txt',
+    ['nemo-xls-to-txt', 'nemo-odf-to-txt'],
     install_dir: join_paths(get_option('prefix'), get_option('bindir')),
     install_mode: 'rwxr-xr-x'
 )
@@ -31,6 +31,7 @@ install_data(
     'mso.nemo_search_helper',
     'mso-ppt.nemo_search_helper',
     'mso-xls.nemo_search_helper',
+    'odf.nemo_search_helper',
     install_dir: join_paths(nemoDataPath, 'search-helpers')
 )
 

search-helpers/nemo-odf-to-txt

@@ -0,0 +1,48 @@
+#!/usr/bin/python3
+
+import sys
+import zipfile
+from pathlib import Path
+from html.parser import HTMLParser
+
+class Parser(HTMLParser):
+    parsed = []
+    get_next_data = False
+    concat_next_data = False
+
+    def handle_starttag(self, tag, attrs):
+        if tag.startswith("text") or tag.startswith("meta"):
+            if tag == "text:s":
+                self.concat_next_data = True
+            self.get_next_data = True
+
+    def handle_endtag(self, tag):
+        pass
+
+    def handle_data(self, data):
+        if self.get_next_data:
+            if data != "\n":
+                if self.concat_next_data:
+                    self.parsed[-1] += " " + data
+                    self.concat_next_data = False
+                else:
+                    self.parsed.append(data.strip())
+                self.get_next_data = False
+
+path = sys.argv[1]
+
+parser = Parser()
+zipfile = zipfile.ZipFile(path)
+files = zipfile.infolist()
+
+for f in files:
+    if f.filename in ("meta.xml", "content.xml"):
+        contents = zipfile.read(f.filename).decode()
+        parser.feed(contents)
+        if len(parser.parsed) == 0:
+            continue
+
+        out_str = ", ".join(parser.parsed)
+        print(f"{Path(f.filename).stem}: {out_str}\n", flush=True, file=sys.stdout)
+        parser.parsed = []
+exit(0)

search-helpers/third-party/libreoffice.nemo_search_helper renamed to search-helpers/odf.nemo_search_helper

@@ -1,5 +1,5 @@
 [Nemo Search Helper]
-TryExec=odt2txt;
-Exec=odt2txt %s
+TryExec=nemo-odf-to-txt;
+Exec=nemo-odf-to-txt %s
 MimeType=application/vnd.oasis.opendocument.text;application/vnd.oasis.opendocument.spreadsheet;application/vnd.oasis.opendocument.presentation;application/vnd.oasis.opendocument.graphics;
-priority=100
+Priority=100

search-helpers/third-party/meson.build

@@ -2,7 +2,6 @@ helpers = [
   'untex.nemo_search_helper',
   'exif.nemo_search_helper',
   'id3.nemo_search_helper',
-  'libreoffice.nemo_search_helper',
   'mso-doc.nemo_search_helper',
   'pdf2txt.nemo_search_helper',
   'pdftotext.nemo_search_helper',