I used a concordance file. It worked fine a few times then the index got corrupted. These tags are left in the document (.fodt in text editor) even when I create a new index WITHOUT a concordance file. How do I remove all of these?
<text:alphabetical-index-mark text:string-value=" " text:key1=“HELLO”/>
<text:alphabetical-index-mark text:string-value=" " text:key1="HELLO: “/>
<text:alphabetical-index-mark text:string-value=” " text:key1="HELLO: : “/>
<text:alphabetical-index-mark text:string-value=” " text:key1=“HELLO: : : “/>
<text:alphabetical-index-mark-start text:id=“IMark94624035441336”/>
<text:span text:style-name=“T3”>HELLO</text:span>
<text:alphabetical-index-mark-end text:id=“IMark94624035441336”/>
This sed command doesn’t seem to do anything:
sed 's/<text:alphabetical-index-mark text:string-value=”([A-Za-z]*)”/>//g’ book.fodt > book2.fodt
There should be a button in the UI: “Remove All Index Entries Real Fast”. This bug has created ~2000 entries, so deleting them one by one in the UI isn’t satisfactory.
EDIT your question (don’t use a comment) to add OS name, LO version and save format (apparently .odt but it is better to state it explicitly). Attached a reduced version of your document (2-3 pages max with the index) and a portion of your concordance file.
At a first glance there is an additional slash before the g. But why would you like to replace with > ?
.
Also I can not see why/how you’ve chosen your search string, so maybe it is better if you don’t use sed to bring more chaos to your xml.
I used these 2 commands to strip the corrupted index tags from the document:
sed ‘s/<text:alphabetical-index-mark text:string-value=" " text:key1="([A-Za-z, ])(: )"/>//g’ new2.fodt > new3.fodt
sed ‘s/<text:alphabetical-index-mark text:string-value=".*"/>//g’ new3.fodt > new4.fodt
The excessive index tags got inserted by a bug in Libre Writer as I was experimenting with using the UI to create an index with and without a concordance file. Each time I created an index it apparently created extra index tags with colons, for no reason.
The original post includes a relevant snippet of the document and explains that the index was corrupt even when not using a concordance file.
After I fixed the problem as described above, the Index (using concordance) looked great. I “updated” it twice–no problem. When I updated it the third time, it got corrupted again. This time, all Index entries appear TWICE in the Index.
Here’s a snippet from the concordance file:
triangle theory;;;;1;1
Aerobie;;;;1;1
Stancil Johnson;Johnson, Dr. Stancil;;;1;1
Here’s a snippet from .fodt:
<text:alphabetical-index-mark text:string-value=" " text:key1=“Aerobie”/>
<text:alphabetical-index-mark-start text:id=“IMark94604660663928”/>Aerobie
<text:alphabetical-index-mark-end text:id=“IMark94604660663928”/>.