From 787794cdd85e7caaa05a902ec580b386020d30cf Mon Sep 17 00:00:00 2001
From: Waldi Ravens <wr86420@gmail.com>
Date: Sun, 28 Jun 2015 19:44:04 +0200
Subject: [PATCH] Updated Serbian OCRFixReplaceList

---
 Dictionaries/srp_OCRFixReplaceList.xml | 61 +++++++++++++-------------
 1 file changed, 30 insertions(+), 31 deletions(-)

diff --git a/Dictionaries/srp_OCRFixReplaceList.xml b/Dictionaries/srp_OCRFixReplaceList.xml
index 1b093202b..1363c6442 100644
--- a/Dictionaries/srp_OCRFixReplaceList.xml
+++ b/Dictionaries/srp_OCRFixReplaceList.xml
@@ -36,12 +36,12 @@
   <EndLines />
   <WholeLines />
   <RegularExpressions>
-    <RegEx find="([a-zčđšž])Ij([a-zčđšž])" replaceWith="$1lj$2" />
-    <RegEx find="([^A-ZČĐŠŽa-zčđšž])Ij(ubav|ud|ut)" replaceWith="$1lj$2" />
+    <RegEx find="(?&lt;=[a-zčđšž])Ij(?=[a-zčđšž])" replaceWith="lj" />
+    <RegEx find="(?&lt;=[^A-ZČĐŠŽa-zčđšž])Iju(?=bav|d|t)" replaceWith="lju" />
     <!-- kad ima razmak izmedju tagova </i> <i> -->
-    <RegEx find="(&gt;) +(&lt;)" replaceWith="$1$2" />
+    <!-- <RegEx find="(&gt;) +(&lt;)" replaceWith="$1$2" /> -->
     <!-- ',"' to '",' -->
-    <RegEx find="(\w),&quot;(\s|$)" replaceWith="$1&quot;,$2" />
+    <RegEx find="(?&lt;=\w),&quot;(?=\s|$)" replaceWith="&quot;," />
     <RegEx find=",\.{3}|\.{3},|\.{2} \." replaceWith="..." />
     <!-- "1 :", "2 :"... "n :" to "n:" -->
     <RegEx find="(\d) +: +(\D)" replaceWith="$1: $2" />
@@ -61,15 +61,14 @@
     <!-- Two or more consecutive '"' to one '"' -->
     <RegEx find="&quot;{2,}" replaceWith="&quot;" />
     <!-- Fix zero and capital 'o' ripping mistakes -->
-    <RegEx find="([0-9])O" replaceWith="${1}0" />
-    <RegEx find="([0-9])\.O" replaceWith="$1.0" />
-    <RegEx find="\b0([A-Za-z])" replaceWith="O$1" />
+    <RegEx find="(?&lt;=[0-9]\.?)O" replaceWith="0" />
+    <RegEx find="\b0(?=[A-ZČĐŠŽa-zčđšž])" replaceWith="O" />
     <!-- Brisanje crte - na početku 1. reda (i kada ima dva reda) -->
     <RegEx find="\A- ?([A-ZČĐŠŽa-zčđšž0-9„'&quot;]|\.{3})" replaceWith="$1" />
     <RegEx find="\A(&lt;[i|b|u]&gt;)- ?" replaceWith="$1" />
     <RegEx find="  - " replaceWith=" -" />
     <!-- Brisanje razmaka iza crte - na početku 2. reda -->
-    <RegEx find="([\n](&lt;[i|b|u]&gt;)?)- ([0-9A-ZČĐŠŽčš„'&quot;&lt;])" replaceWith="$1-$3" />
+    <RegEx find="(?&lt;=\n(&lt;[i|b|u]&gt;)?)- (?=[A-ZČĐŠŽčš0-9„'&quot;&lt;])" replaceWith="-" />
     <!-- Korigovanje crte - kad je u sredini prvog reda -->
     <RegEx find="([.!?&quot;&gt;]) - ([A-ZČĐŠŽčš'&quot;&lt;])" replaceWith="$1 -$2" />
     <!-- Zatvoren tag pa razmak poslije crtice -->
@@ -80,37 +79,37 @@
     <RegEx find="\(- ([A-ZČĐŠŽčš„'&quot;])" replaceWith="(-$1" />
     <!-- Smart space after dot -->
     <!-- osim kad je zadnje t (rijec kolt) -->
-    <RegEx find="([a-su-zá-úñä-ü])\.([^\s\n().:?!*^“”'&quot;&lt;])" replaceWith="$1. $2" />
+    <RegEx find="(?&lt;=[a-su-zá-úñä-ü])\.(?=[^\s\n().:?!*^“”'&quot;&lt;])" replaceWith=". " />
     <!-- Oznaka za kalibar. Npr. "Colt .45" -->
     <!-- Da bi radilo, da bi ovaj razmak bio dozvoljen, u SW idite Alt+I i odcekirajte "Razmaci ispred tacke" -->
-    <RegEx find="t\.([0-9][0-9])" replaceWith="t .$1" />
+    <RegEx find="t\.(?=[0-9]{2})" replaceWith="t ." />
     <!-- Joey(j)a -->
-    <RegEx find="([A-Z][a-z])eyj([a-z])" replaceWith="$1ey$2" />
+    <RegEx find="(?&lt;=\b[A-Z][a-z])eyj(?=[a-z])" replaceWith="ey" />
     <!-- Sređuje zarez sa razmakom -->
-    <RegEx find="([A-ZČĐŠŽa-zčđšžá-úñä-ü&quot;]),([^\s\n(),?!“&lt;])" replaceWith="$1, $2" />
-    <RegEx find=" , " replaceWith=", " />
-    <RegEx find=" ,([A-ZČĐŠŽa-zčđšž])" replaceWith=", $1" />
-    <RegEx find=" ,$" replaceWith="," />
+    <RegEx find="(?&lt;=[A-ZČĐŠŽa-zčđšžá-úñä-ü&quot;]),(?=[^\s(),?!“&lt;])" replaceWith=", " />
+    <RegEx find=" +,(?=[A-ZČĐŠŽa-zčđšž])" replaceWith=", " />
+    <RegEx find=" +, +" replaceWith=", " />
+    <RegEx find=" +,$" replaceWith="," />
     <RegEx find="([?!])-" replaceWith="$1 -" />
     <!-- Space after last of some consecutive dots (eg. "...") -->
-    <RegEx find="([a-zčđšž])(\.{3}|!)([a-zčđšž])" replaceWith="$1$2 $3" />
+    <RegEx find="(?&lt;=[a-zčđšž])(\.{3}|!)(?=[a-zčđšž])" replaceWith="$1 " />
     <!-- Delete space after "..." that is at the beginning of the line. You may delete this line if you don't like it -->
     <!-- <RegEx find="^\.{3} +" replaceWith="..." /> -->
     <!-- "tekst ... tekst" mijenja u "tekst... tekst" -->
-    <RegEx find="([A-ZČĐŠŽa-zčđšž]) \.{3} " replaceWith="$1... " />
-    <RegEx find="(\S)\. &quot;" replaceWith="$1.&quot;" />
-    <RegEx find="&quot; \." replaceWith="&quot;." />
-    <RegEx find="(\S\.{3}) &quot;(\s|$)" replaceWith="$1&quot;$2" />
-    <RegEx find=" \.{3}$" replaceWith="..." />
-    <RegEx find="([a-zčđšž])( \.{3}|\.{2}$)" replaceWith="$1..." />
+    <RegEx find="(?&lt;=[A-ZČĐŠŽa-zčđšž]) +\.{3} +" replaceWith="... " />
+    <RegEx find="(?&lt;=\S)\. +&quot;" replaceWith=".&quot;" />
+    <RegEx find="&quot; +\." replaceWith="&quot;." />
+    <RegEx find="(?&lt;=\S\.{3}) +&quot;(?=\s|$)" replaceWith="&quot;" />
+    <RegEx find=" +\.{3}$" replaceWith="..." />
+    <RegEx find="(?&lt;=[a-zčđšž])(?: +\.{3}|\.{2}$)" replaceWith="..." />
     <!-- Razmak ispred zagrade -->
-    <RegEx find="([A-ZČĐŠŽa-zčđšž])\(" replaceWith="$1 (" />
+    <RegEx find="(?&lt;=[A-ZČĐŠŽa-zčđšž])\(" replaceWith=" (" />
     <!-- Razmak iza upitnika -->
-    <RegEx find="\?([A-ZČĐŠŽčš])" replaceWith="? $1" />
-    <RegEx find="(^|&gt;)\.{3} ([A-ZČĐŠŽčš])" replaceWith="$1...$2" />
+    <RegEx find="\?(?=[A-ZČĐŠŽčš])" replaceWith="? " />
+    <RegEx find="(?&lt;=^|&gt;)\.{3} +(?=[A-ZČĐŠŽčš])" replaceWith="..." />
     <!-- Brise ... kad je na poc. reda "... -->
-    <RegEx find="^&quot;\.{3} " replaceWith="&quot;" />
-    <RegEx find="([0-9])\$" replaceWith="$1 $" />
+    <RegEx find="^&quot;\.{3} +" replaceWith="&quot;" />
+    <RegEx find="(?&lt;=[0-9])\$" replaceWith=" $" />
     <!-- ti š -> t š by Strider -->
     <!-- Zamijeni sva "**ti šu*" s "**t šu*" i "**ti še*" s "**t še*" -->
     <!-- <RegEx find="([a-z])ti (š+[eu])" replaceWith="$1t $2" /> -->
@@ -120,19 +119,19 @@
     <RegEx find="^(&lt;[i|b|u]&gt;) +" replaceWith="$1" />
     <RegEx find="^\.{2} +" replaceWith="..." />
     <!-- Razmak ? "</i> -->
-    <RegEx find="([.?!]) +(&quot;&lt;)" replaceWith="$1$3" />
+    <RegEx find="([.?!]) +(&quot;&lt;)" replaceWith="$1$2" />
     <!-- Bez razmaka kod Npr.: -->
-    <RegEx find="([Nn])pr\. +:" replaceWith="$1pr.:" />
+    <RegEx find="(?&lt;=[Nn]pr\.) *: *" replaceWith=": " />
     <RegEx find="\. ," replaceWith=".," />
     <RegEx find="([?!])\." replaceWith="$1" />
     <!-- Da ne kvari potpise sa ..:: -->
     <RegEx find="\.{3}::" replaceWith="..::" />
     <RegEx find="::\.{3}" replaceWith="::.." />
-    <RegEx find="\.{2} ::" replaceWith="..::" />
+    <RegEx find="\.{2} +::" replaceWith="..::" />
     <!-- Skracenice bez razmaka -->
     <RegEx find="d\. o\.o\." replaceWith="d.o.o." />
     <!-- Kad red počinje sa ...pa malo slovo -->
     <!-- <RegEx find="^\.{3}([a-zčđšž&quot;&lt;])" replaceWith="$1" /> -->
-    <!-- <RegEx find=" ([.?!])" replaceWith="$1" /> -->
+    <!-- <RegEx find=" +([.?!])" replaceWith="$1" /> -->
   </RegularExpressions>
 </OCRFixReplaceList>
\ No newline at end of file