Колдонуучу:KETC Parser for WikiMediaExport - Түзөтүүлөр тарыхы

Kadyrm: /* AI text processing tools overview */

2026-08-28T10:48:00Z

AI text processing tools overview

← Мурунку нускасы		10:48, 28 Август (Баш оона) 2026 -деги абалы
5 сап:		5 сап:
	= <font color="red"> AI text processing tools overview </font> =		= <font color="red"> AI text processing tools overview </font> =
	Template usage test:		Template usage test:
	{{NestedList\|Introduction to MediaWiki\|author=Jane Doe}}		{{NestedList\|Introduction to MediaWiki\|author=Jane Doe\|year = 1990}}

	== '''UP: which AI tool would you recommend to process a formatted text stored in docx or odt format''' ==		== '''UP: which AI tool would you recommend to process a formatted text stored in docx or odt format''' ==

Kadyrm: /* AI text processing tools overview */

2026-08-28T10:44:39Z

AI text processing tools overview

← Мурунку нускасы		10:44, 28 Август (Баш оона) 2026 -деги абалы
4 сап:		4 сап:

	= <font color="red"> AI text processing tools overview </font> =		= <font color="red"> AI text processing tools overview </font> =
			Template usage test:
			{{NestedList\|Introduction to MediaWiki\|author=Jane Doe}}

	== '''UP: which AI tool would you recommend to process a formatted text stored in docx or odt format''' ==		== '''UP: which AI tool would you recommend to process a formatted text stored in docx or odt format''' ==
	Response: If you are referring to custom, proprietary, or highly specific data structures (like internal log layouts or formatted configuration files), '''ChatGPT Plus (GPT-4o)''' or '''Claude 3.5 Sonnet''' are the best tools.		Response: If you are referring to custom, proprietary, or highly specific data structures (like internal log layouts or formatted configuration files), '''ChatGPT Plus (GPT-4o)''' or '''Claude 3.5 Sonnet''' are the best tools.

Kadyrm: /* UP: which AI tool would you recommend to process a formatted text in docx or odt format */

2026-08-28T10:04:52Z

UP: which AI tool would you recommend to process a formatted text in docx or odt format

← Мурунку нускасы		10:04, 28 Август (Баш оона) 2026 -деги абалы
4 сап:		4 сап:

	= <font color="red"> AI text processing tools overview </font> =		= <font color="red"> AI text processing tools overview </font> =
	== '''UP: which AI tool would you recommend to process a formatted text in docx or odt format''' ==		== '''UP: which AI tool would you recommend to process a formatted text stored in docx or odt format''' ==
	Response: If you are referring to custom, proprietary, or highly specific data structures (like internal log layouts or formatted configuration files), '''ChatGPT Plus (GPT-4o)''' or '''Claude 3.5 Sonnet''' are the best tools.		Response: If you are referring to custom, proprietary, or highly specific data structures (like internal log layouts or formatted configuration files), '''ChatGPT Plus (GPT-4o)''' or '''Claude 3.5 Sonnet''' are the best tools.

Kadyrm: /* Classification for Parsing */

2026-08-28T09:29:56Z

Classification for Parsing

← Мурунку нускасы		09:29, 28 Август (Баш оона) 2026 -деги абалы
4 460 сап:		4 460 сап:
	----		----

	= <span style="color: red;">~~Classification for~~ Parsing</span> =		= <span style="color: red;">MediaWiki Export Schema led Parsing</span> =
	== '''UP: there are 3 types of p/strong elements: 1) containing all capital unicode chars like <nowiki><strong>ФЫАЫФА</strong></nowiki>. this is actually an article title; 2) first letter capital like <nowiki><strong>Авааф фыаа</strong></nowiki>. This is a section header inside an article; 3) all letters small like <nowiki><strong>выфафыа</strong></nowiki>. This is just an inline bold text having no structural load. As a long term goal I need to classify all of them in my html document. As a specific quetion: can I use regex pattern inside XPath expression to match all capitals and first letter capital text patterns?''' ==		== '''UP: Classification for Parsing: there are 3 types of p/strong elements: 1) containing all capital unicode chars like <nowiki><strong>ФЫАЫФА</strong></nowiki>. this is actually an article title; 2) first letter capital like <nowiki><strong>Авааф фыаа</strong></nowiki>. This is a section header inside an article; 3) all letters small like <nowiki><strong>выфафыа</strong></nowiki>. This is just an inline bold text having no structural load. As a long term goal I need to classify all of them in my html document. As a specific quetion: can I use regex pattern inside XPath expression to match all capitals and first letter capital text patterns?''' ==
	Response: '''No, you cannot use native regular expressions directly inside standard XPath 1.0 expressions.'''		Response: '''No, you cannot use native regular expressions directly inside standard XPath 1.0 expressions.'''

Kadyrm: /* UP: there are 3 types of p/strong elements: 1) containing all capital unicode chars like ФЫАЫФА. this is actually an article title; 2) first letter capital like Авааф фыаа. This is a section header inside an article; 3) all letters small like выфафыа. This is just an inline bold text having no structural load. As a long term goal I need to classify all of them in my html document. As a specific quetion: can I use regex pattern inside XPath expression to match all capitals and first letter...

2026-08-28T09:26:08Z

/* UP: there are 3 types of p/strong elements: 1) containing all capital unicode chars like ФЫАЫФА. this is actually an article title; 2) first letter capital like Авааф фыаа. This is a section header inside an article; 3) all letters small like выфафыа. This is just an inline bold text having no structural load. As a long term goal I need to classify all of them in my html document. As a specific quetion: can I use regex pattern inside XPath expression to match all capitals and first letter...

← Мурунку нускасы		09:26, 28 Август (Баш оона) 2026 -деги абалы
4 460 сап:		4 460 сап:
	----		----

			= <span style="color: red;">Classification for Parsing</span> =
	== '''UP: there are 3 types of p/strong elements: 1) containing all capital unicode chars like <nowiki><strong>ФЫАЫФА</strong></nowiki>. this is actually an article title; 2) first letter capital like <nowiki><strong>Авааф фыаа</strong></nowiki>. This is a section header inside an article; 3) all letters small like <nowiki><strong>выфафыа</strong></nowiki>. This is just an inline bold text having no structural load. As a long term goal I need to classify all of them in my html document. As a specific quetion: can I use regex pattern inside XPath expression to match all capitals and first letter capital text patterns?''' ==		== '''UP: there are 3 types of p/strong elements: 1) containing all capital unicode chars like <nowiki><strong>ФЫАЫФА</strong></nowiki>. this is actually an article title; 2) first letter capital like <nowiki><strong>Авааф фыаа</strong></nowiki>. This is a section header inside an article; 3) all letters small like <nowiki><strong>выфафыа</strong></nowiki>. This is just an inline bold text having no structural load. As a long term goal I need to classify all of them in my html document. As a specific quetion: can I use regex pattern inside XPath expression to match all capitals and first letter capital text patterns?''' ==
	Response: '''No, you cannot use native regular expressions directly inside standard XPath 1.0 expressions.'''		Response: '''No, you cannot use native regular expressions directly inside standard XPath 1.0 expressions.'''

Kadyrm: /* UP: still removes other imgs. #problematic; */

2026-08-28T09:22:59Z

UP: still removes other imgs. #problematic;

← Мурунку нускасы		09:22, 28 Август (Баш оона) 2026 -деги абалы
3 745 сап:		3 745 сап:
	----		----

	== '''UP: still removes other imgs. #problematic;''' ==		== '''UP: still removes other imgs. <span style="color: red;">#problematic;</span>''' ==
	Response: The issue happens because absolute XPath axes like following-sibling::div[1] can behave unexpectedly depending on whitespace/text nodes in HTML, accidentally matching paragraphs that aren't ''strictly'' adjacent to your target class.		Response: The issue happens because absolute XPath axes like following-sibling::div[1] can behave unexpectedly depending on whitespace/text nodes in HTML, accidentally matching paragraphs that aren't ''strictly'' adjacent to your target class.

Kadyrm: /* UP: still removes other imgs */

2026-08-28T09:22:39Z

UP: still removes other imgs

← Мурунку нускасы		09:22, 28 Август (Баш оона) 2026 -деги абалы
3 745 сап:		3 745 сап:
	----		----

	== '''UP: still removes other imgs''' ==		== '''UP: still removes other imgs. #problematic;''' ==
	Response: The issue happens because absolute XPath axes like following-sibling::div[1] can behave unexpectedly depending on whitespace/text nodes in HTML, accidentally matching paragraphs that aren't ''strictly'' adjacent to your target class.		Response: The issue happens because absolute XPath axes like following-sibling::div[1] can behave unexpectedly depending on whitespace/text nodes in HTML, accidentally matching paragraphs that aren't ''strictly'' adjacent to your target class.

Kadyrm: /* Gamechanger: Mammoth lib */

2026-08-28T09:18:14Z

Gamechanger: Mammoth lib

← Мурунку нускасы		09:18, 28 Август (Баш оона) 2026 -деги абалы
2 624 сап:		2 624 сап:
	----		----

	= <span style="color: red;">~~Gamechanger:~~ Mammoth lib</span> =		= <span style="color: red;">Application of Mammoth lib</span> =

	== '''UP: will mammoth produce me a sanitized html of the docx file keeping formatting intact?''' ==		== '''UP: will mammoth produce me a sanitized html of the docx file keeping formatting intact?''' ==

Kadyrm: /* Using Python (mammoth) */

2026-08-28T09:01:16Z

Using Python (mammoth)

← Мурунку нускасы		09:01, 28 Август (Баш оона) 2026 -деги абалы
2 059 сап:		2 059 сап:
	* '''Why:''' It preserves the "Heading 1" or "Bold" tags from your Word file and converts them into Markdown # or ** syntax.		* '''Why:''' It preserves the "Heading 1" or "Bold" tags from your Word file and converts them into Markdown # or ** syntax.

	==== '''Using Python (mammoth)''' ====		==== <span style="color: red;">'''#knowledge; #critical;</span> Using Python (mammoth)''' ====
	If you want to keep this inside your Python project, use the mammoth library. Unlike other libraries, mammoth is specifically designed to convert DOCX to clean HTML or Markdown by mapping styles.		If you want to keep this inside your Python project, use the mammoth library. Unlike other libraries, mammoth is specifically designed to convert DOCX to clean HTML or Markdown by mapping styles.

Kadyrm: /* Using pandoc (The Gold Standard) */

2026-08-28T08:59:31Z

Using pandoc (The Gold Standard)

← Мурунку нускасы		08:59, 28 Август (Баш оона) 2026 -деги абалы
2 052 сап:		2 052 сап:
	You do not need to do this manually. You can use Python to "translate" the DOCX styles into Markdown syntax automatically.		You do not need to do this manually. You can use Python to "translate" the DOCX styles into Markdown syntax automatically.

	==== '''Using pandoc (The Gold Standard)''' ====
			==== '''#knowledge: Using pandoc (The Gold Standard)''' ====
	Pandoc is a command-line tool that is considered the "Swiss-army knife" of document conversion. It is incredibly good at converting DOCX styles directly into clean Markdown.		Pandoc is a command-line tool that is considered the "Swiss-army knife" of document conversion. It is incredibly good at converting DOCX styles directly into clean Markdown.