User:Camomiletea/Regexes

Regexes I found useful

Search	Replace	Function
^-{5}File: ([^.]+)\.(pn|jp)g.*	-----File: $1.$2g-----	Strip proofer names
([,.;:'"!?])-\b	$1--	Find dashes after punctuation that should be em-dashes
\b-([.,;:'"!?])	--$1	Find dashes before punctuation that should be em-dashes
Mrs?[^\.]	(none)	Mr vs. Mr.
\n\n\n	(none)	Check chapter/section spacing
---	(none)	Check dashes
(\d+)--(\d+)	$1-$2	Correct em-dashes between numbers to dashes
[\x{0100}-\x{ffffff}]		Check for Unicode characters
[\x7F-\x{0100}]		Check for non-ASCII characters
</?i>	_	Converting italics in text version
<(/?)i>	<$1em>	Converting italics in HTML, if desired
<p>(\P{IsLower}+)</p>	<h3>$1</h3>	Convert all upper-case one-line paragraphs to headings in HTML
<sc>((.|\n)+?)</sc>	\U$1\E	Converting small-caps in text version
([A-Z]\.) ([A-Z]\.)	$1&nbsp;$2	convert spaces in initials to no-breaking spaces in HTML
"pagenum"><a name="Page_(\d+)" id="Page_\1">\[Pg \1\]</a>	"pagenum" title="Page&nbsp;$1">&nbsp;<a name="Page_$1" id="Page_$1"></a>	Convert auto-generated pagenum output in HTML why?

Creating Guiguts-style dictionary from word lists

Ensure that the word list is as you want it: it is case-sensitive!

1. First, you must escape all apostrophes; i.e. find ' and replace with \'
2. Use the following regex: find ^(.*)$ and replace with '$1' => '',
3. At the beginning add on a separate line: %projectdict = (
4. At the end, remove the final comma, and on a separate line add: );

Missing periods

• t[\s\n]\p{IsUpper}
  • It seems that early 20th century typesetters often squashed the "t." closely enough together for the OCR to think it's just a "t".
• Missing periods after Roman Numerals
• Missing period in "&c.," when followed by a comma
• ".," often becomes just a comma, or a semicolon
• per cent. - in books where a period is used in this phrase (don't make it a comma)
• [XIVLC]L
  • find a Roman numeral ending in L, which should be I

See Also

Notes on processing old Russian orthography