Indentation issue when converting to html

Hi Team,
We are using libreoffice 4.3 to convert word document to HTML. The document generated by libre office has indentation issues. In the word document, we have many p tags as ordered list which after conversion adds inline css.

Here is an example of the code snippet of the converted html document:

<p align="left" style="margin-left: 0.25in; text-indent: -0.25in; margin-bottom: 0.17in; line-height: 0.17in; widows: 2; orphans: 2">
<a name="5c188da9-cdb8-406b-9809-dd4a0316dba2-1"></a>
<font face="Arial, serif">
<font size="2" style="font-size: 11pt">
1. Some Text Here

Please look into this asap.

(Edited 2018-01-11 by @sakshi:slight_smile: Now there is a demo file: SampleDocument_IndentIssue.docx

(Edited slightly for better readability by @Lupp )

Look into what? “1. Some Text Here”? and look who? users using this peer-to-peer help forum?

If you found a bug, then you should file a bug report; and as stated in this forum’s rules (available on the forum’s main page as “How to use this Ask site”), bug reports are off-topic here.

Sorry! There is no “team” here, but independent volunteers.
Sorry again: I don’t now enough about your issue to answer the question, but:
Your issue is specific to ‘Writer’. Please edit your question and add the tag writer. You may remove common instead.

Oh! Just noticed the version there… 4.3! And the asker wants someone to “look into this ASAP”! Sigh…

@sakshi: You recently updated your question. (I cannot find the differences yet.) But you did not regard the comment by @mikekaganski here above. To be clear:
Your version 4.3 has reached end of development long ago. The same with 4.4, and there never was a 4.5. The series of versions for which still bug fixing is done is 5.4. The next “fresh” version 6.0 is to come soon. is presently in Beta2 state

@sakshi: I still can’t find what changs you made by editing.
And I still don’t know what indentation issues there are.


Can the issue of the indentation can be resolved or is handled in the version 5.4.

Sakshi Singh

Did you read my answer below?
Without better information there will not be any progress in this thread. As far as I can see nobody can even tell if there actually is an issue that might need a fix.
I perssonally cannot find the information needed for trying to find out if there is a related bug report.


We can not share the word document or the screenshot of the resultant html due to some security issues.
If we were able to show it the issue can be better explained thus can we have a discussion over call if possible so that I can explain the issue.

Kindly let us know in case of any concerns.


I will grant the “karma” needed for attaching documents. (I am not much interested in screenshots. To check some html code for the way it gets rendered I can use my own browser.)
If y documnt type is rejected you may append a fake extension.

@Lupp: the OP doesn’t want to attach a document here (referring to some “security issues”). As if one couldn’t create a sample with the issue, but without sensitive information. :slight_smile:


Where I can attach the document here. I have created a sample document describing the issue.


@sakshi: edit your question above, and use an “attachment (Ctrl+F)” button to attach the file, as @Lupp already has given you required access rights.


I have uploaded the document referring the issue.
Kindly check and let me know in case of any concerns.

Sakshi Singh

For reference: tdf#114591


Is the attached document was helpful for the understanding the issue. Can this issue be resolved using new version or with the same version used.

Sakshi Singh

Quoting @sakshi: “Is the attached document was helpful…”
May I assume there was a ‘typo’, and you wanted to start wit “If…”?

The issue is clear. It exists in all versions, including current master. It also happens with both HTML output filters (available in Save As and Export).

Thank you for filing a bug report for this, and providing a sample (finally!). We would be also thankful for any patches. Other than that, there can be no guarantees as of when (and if) it will be improved. You can look for certified developers and companies providing L3 support, if you need.

@mikekaganski: Your comment above looks as if you can confirm a bug concerning the creation of html code by LibO.
As I also tried to analyse the attached example, I came to the conclusion that the hoped-for result was not to be expected. It would require LibO to autoamtically convert a sequence of paragraphs formatted to a very special style to a numbered list in html. See the amendment to my answer (to come) for mor detail.

Of course, the hoped-for result isn’t got with LO export. That’s the OP’s issue, and I see it in the LO output. As I see it, OP didn’t post a hoped-for code, but the actual result of export that OP is unhappy with.

I do see it as a minor issue, because LO isn’t a HTML editor, and some loose of formatting is expected. The wanted behavior would require us to generate bloated HTML, for little gain.