[scribus] Paragraph Separator and Newline Both Showing as \r in Python

[Matt Miller]

On Fri, Dec 31, 2021, at 15:59, Gregory Pittman wrote:
> On 12/31/21 14:19, Matt Miller wrote:
>> On Fri, Dec 31, 2021, at 07:11, Gregory Pittman wrote:
>>> On 12/30/21 20:49, Matt Miller wrote:
>>>> I'm loading a text frame from a utf-8 encoded text file, and within my Scribus Python code I want to search for the standard newline character, ascii value 10. When I see an ascii 10 as the line separator I want to apply a special paragraph style to the following paragraph. Most paragraphs end with the Unicode paragraph separator character, \u2029, and in those cases the default paragraph style is fine.
>>>>
>>>> My problem is that both these types of characters are matching '\r' when I use re.search in python. also, if I select either line separator character, then do getText(), I get a '\r' no matter what. I've confirmed that my file encoding is utf-8. What am I missing? How can I search for a simple '\n' character?
>>>
>>> Hi Matt,
>>>
>>> You don't say what OS you're using.
>> 
>> Linux
>> 
>>> Maybe running dos2unix on the text would help.
>> 
>> Well, my ideal workflow is one where the contents of the text file instruct Scribus what paragraph and character styles to use throughout the document. So, I'm careful to put exactly the characters I want into the text file. I've been largely successful, but now I've run into a case where it seems Scribus (or Python) is losing information when I load the file. From inside Scribus I can't distinguish between a Unicode paragraph separator, \u2029, and a simple line feed, \u000A.
>> 
>> I'm able to open my text file from the Python console and dump it out to see that newline is displayed as "\n" and a paragraph separator is displayed as "\u2029." So, I'm suspecting the problem is with Scribus, or how I'm using it. I'm loading the file using insertHtmlText(), but I get the same bad behavior from the GUI when I do "Content | Get Text..." and load the file manually.
>> 
>> I've attached a text file that shows the problem. If you run the following in the scripter console from a document with a text frame named "Text1" you should see the problem:
>> 
>
> When I save this file then open it with kwrite, then when I do a 
> Replace operation trying to switch \n to <p> (an arbitrary choice), I 
> see the replacement happen at the end of the 2nd, 3rd, 4th, and 6th 
> sentences.

I duplicated that behavior with kwrite. It seems only the 2nd and 6th sentences should have been replaced, since only those have an actual \u000A. But, I guess each app does some of its own interpretation as to what '\n' means. I suppose that's similar to what Scribus is doing when it decides that \u2029 and \u000A should both appear as '\r' in the Python string.

So, I changed my special marker character that I expect in my text files from a line feed to a zero-width space. That character isn't as semantically meaningful to me for this purpose as a line feed, but it works.

Thanks for the input.

> Greg
>
>
> ___
> Scribus Mailing List: scribus at lists.scribus.net
> Edit your options or unsubscribe:
> http://lists.scribus.net/mailman/listinfo/scribus
> See also:
> http://wiki.scribus.net
> http://forums.scribus.net

-- 

  Matt Miller
  mailto:matt.miller at fastmail.com