Add automated tests

.github/workflows/test.yml

@@ -16,3 +16,21 @@ jobs:
         chmod +x /usr/local/bin/tidy
     - name: Validate HTML
       run: tidy -errors -quiet --drop-empty-elements no regex.html
+
+  test_sed:
+    strategy:
+      matrix:
+        os: [ubuntu-latest, macos-latest]
+      fail-fast: false
+    runs-on: ${{ matrix.os }}
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run sed tests
+      run: tests/test_sed.sh
+
+  set_vim:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run Vim tests
+      run: tests/test_vim.sh

regex.html

@@ -82,7 +82,7 @@ <h1>Syntax</h1>
 
 <table class="table table-bordered table-striped">
   <thead>
-    <tr><th>What</th><th><a href="http://perldoc.perl.org/perlre.html">Perl</a>/PCRE</th><th><a href="https://docs.python.org/library/re.html">Python's <code>re</code></a></th><th>POSIX (BRE)</th><th>POSIX extended (ERE)</th><th>Vim</th></tr>
+    <tr><th>What</th><th><a href="http://perldoc.perl.org/perlre.html">Perl</a>/PCRE</th><th><a href="https://docs.python.org/library/re.html">Python's <code>re</code></a></th><th>POSIX (BRE)</th><th>POSIX extended (ERE)</th><th>Vim (with <code>:set magic</code>)</th></tr>
   </thead>
   <tbody id="syntax-basics" class="subgroup">
     <tr><th colspan="6">Basics</th></tr>
@@ -99,7 +99,7 @@ <h1>Syntax</h1>
     <tr><td>Any character (including newline)</td><td class="na"></td><td class="na"></td><td class="na"></td><td class="na"></td><td><code>\_.</code></td></tr>
     <tr><td>Match a "word" character (alphanumeric plus <code>_</code>)</td><td><code>\w</code> <code>[[:word:]]</code></td><td><code>\w</code></td><td><code>\w</code></td><td><code>\w</code></td><td><code>\w</code></td></tr>
     <tr><td>Case</td><td><code>[[:upper:]]</code> / <code>[[:lower:]]</code></td><td class="na"></td><td><code>[[:upper:]]</code> / <code>[[:lower:]]</code></td><td><code>[[:upper:]]</code> / <code>[[:lower:]]</code></td><td><code>\u</code> <code>[[:upper:]]</code> / <code>\l</code> <code>[[:lower:]]</code></td></tr>
-    <tr><td>Match a non-"word" character</td><td><code>\W</code></td><td><code>\W</code></td><td class="na"></td><td class="na"></td><td><code>\W</code></td></tr>
+    <tr><td>Match a non-"word" character</td><td><code>\W</code></td><td><code>\W</code></td><td><code>\W</code></td><td><code>\W</code></td><td><code>\W</code></td></tr>
     <tr><td>Match a whitespace character (except newline)</td><td class="na"></td><td class="na"></td><td><code>\s</code> <code>[[:space:]]</code></td><td><code>\s</code> <code>[[:space:]]</code></td><td><code>\s</code> <code>[[:space:]]</code></td></tr>
     <tr><td>Whitespace including newline</td><td><code>\s</code> <code>[[:space:]]</code></td><td><code>\s</code></td><td class="na"></td><td class="na"></td><td><code>\_s</code></td></tr>
     <tr><td>Match a non-whitespace character</td><td><code>\S</code></td><td><code>\S</code></td><td><code>[^[:space:]]</code></td><td><code>[^[:space:]]</code></td><td><code>\S</code> <code>[^[:space:]]</code></td></tr>
@@ -153,7 +153,8 @@ <h1>Syntax</h1>
   <tbody id="syntax-other" class="subgroup">
     <tr><th colspan="6">Other</th></tr>
     <tr><td>Independent non-backtracking pattern</td><td><code>(?&gt;...)</code></td><td class="na"></td><td class="na"></td><td class="na"></td><td><code>\(...\)\@&gt;</code></td></tr>
-    <tr><td>Make case-sensitive/insensitive</td><td><code>(?i)</code> / <code>(?-i)</code></td><td><code>(?i)</code> / <code>(?-i)</code></td><td class="na"></td><td class="na"></td><td><code>\c</code> / <code>\C</code></td></tr>
+    <tr><td>Make case-insensitive</td><td><code>(?i)</code></td><td><code>(?i)</code></td><td class="na"></td><td class="na"></td><td><code>\c</code></td></tr>
+    <tr><td>Make case-sensitive</td><td><code>(?-i)</code></td><td class="na"></td><td class="na"></td><td class="na"></td><td><code>\C</code></td></tr>
   </tbody>
 </table>
 </section>

tests/test_sed.sh

@@ -0,0 +1,296 @@
+#!/bin/bash
+
+export LC_ALL=en_US.UTF-8
+
+check_(){
+    RESULT="$(echo "$4" | sed "$1" "$5")"
+    if [ "$RESULT" != "$6" ]; then
+        echo "Test failed: $3 ($2)" >&2
+        echo "----- expected -----
+$6
+----- returned -----"
+        echo "$RESULT"
+        exit 1
+    fi
+}
+
+# Check BRE
+checkb(){
+    check_ "-e" "BRE" "$@"
+}
+# Check ERE
+checke(){
+    check_ "-E" "ERE" "$@"
+}
+# Check both
+check(){
+    checkb "$@"
+    checke "$@"
+}
+
+fail_(){
+    if echo | sed "$1" "$4" &>/dev/null; then
+        echo "Test didn't fail as expected: $3 ($2)" >&2
+        exit 1
+    fi
+}
+
+# Fail check BRE
+failb(){
+    fail_ "-e" "BRE" "$@"
+}
+# Fail check ERE
+faile(){
+    fail_ "-E" "ERE" "$@"
+}
+# Fail check both
+fail(){
+    failb "$@"
+    faile "$@"
+}
+
+# Basics
+
+check 'Custom character class' \
+'adn
+aqv' 's/[b-eq]/x/g' \
+'axn
+axv'
+
+check 'Negated custom character class' \
+'abcdefgh' \
+'s/[^b-dg-l]/x/g' \
+'xbcdxxgh'
+
+check 'Backslash not special in class' \
+"a]\\b
+a]\\b" \
+"1s/[\\m-p]/x/g
+2s/[]]/x/g" \
+"a]xb
+ax\\b"
+
+check 'Ranges' \
+'a-e-i
+a-e-i' \
+'1s/[d-f]/x/g
+2s/[d-f-]/x/g' \
+'a-x-i
+axxxi'
+
+checkb 'Alternation' \
+'acd' \
+'s/b\|c/x/g' \
+'axd'
+checke 'Alternation' \
+'acd' \
+'s/b|c/x/g' \
+'axd'
+
+check 'Escaped character' \
+'abc' \
+"s/\\061\\x62\\x(99)/x/g" \
+'abc'
+
+# Charater classes
+
+check 'Any character' \
+'a ;d
+efg' \
+'s/./x/g' \
+'xxxx
+xxx'
+
+check 'Word character' \
+'hello w0r|_d!' \
+'s/\w/x/g' \
+'xxxxx xxx|xx!'
+
+fail 'Word class' \
+'s/[[:word:]]/x/g'
+
+check 'Upper case' \
+'Hell0 W0r|_D' \
+'s/[[:upper:]]/X/g' \
+'Xell0 X0r|_X'
+
+check 'Lower case' \
+'Hell0 W0r|_D' \
+'s/[[:lower:]]/x/g' \
+'Hxxx0 W0x|_D'
+
+check 'Whitespace' \
+'Hello world	!' \
+'s/\s/_/g' \
+'Hello_world_!'
+
+check 'Whitespace' \
+'Hello world	!' \
+'s/[[:space:]]/_/g' \
+'Hello_world_!'
+
+check 'Non-whitespace' \
+'Hello world	!' \
+'s/[^[:space:]]/x/g' \
+'xxxxx xxxxx	x'
+
+check 'Digit' \
+'H3ll0 W0r|_D' \
+'s/[[:digit:]]/+/g' \
+'H+ll+ W+r|_D'
+
+check 'Hexadecimal digit' \
+'H3ll0 W0r|_D' \
+'s/[[:xdigit:]]/+/g' \
+'H+ll+ W+r|_+'
+
+fail 'Octal digit' \
+'s/[[:odigit:]]/x/g'
+
+check 'Punctuation' \
+'+- 01: hello, world! ;) -+' \
+'s/[[:punct:]]/_/g' \
+'__ 01_ hello_ world_ __ __'
+
+check 'Alphabetical characters' \
+'+- 01: hello, world! ;) -+' \
+'s/[[:alpha:]]/x/g' \
+'+- 01: xxxxx, xxxxx! ;) -+'
+
+check 'Alphanumerical characters' \
+'+- 01: hello, world! ;) -+' \
+'s/[[:alnum:]]/x/g' \
+'+- xx: xxxxx, xxxxx! ;) -+'
+
+fail 'ASCII class' \
+'s/[[:ascii:]]/x/g'
+
+check 'Character equivalents' \
+'Rémi est prêt' \
+'s/[[=e=]]/_/g' \
+'R_mi _st pr_t'
+
+check 'Word boundary' \
+'Hello, world' \
+'s/o\b/x/g' \
+'Hellx, world'
+
+check 'Not word boundary' \
+'Hello, world' \
+'s/o\B/x/g' \
+'Hello, wxrld'
+
+check 'Begining of line' \
+'testing tests' \
+'s/^t/r/g' \
+'resting tests'
+
+check 'End of line' \
+'testing tests' \
+'s/s$/x/g' \
+'testing testx'
+
+# Captures and groups
+
+checkb 'Capturing group' \
+'Name is Remi!' \
+'s/^.*is \(.*\)!$/\1/' \
+'Remi'
+checke 'Capturing group' \
+'Name is Remi!' \
+'s/^.*is (.*)!$/\1/' \
+'Remi'
+
+checkb 'Non-capturing parentheses' \
+'Some (dumb)text' \
+'s/(.*)//g' \
+'Some text'
+checke 'Non-capturing parentheses' \
+'Some (dumb)text' \
+'s/\(.*\)//g' \
+'Some text'
+
+checkb 'Backreference' \
+'ab be cd cc df' \
+'s/\([a-z]\)\1/xx/g' \
+'ab be cd xx df'
+checke 'Backreference' \
+'ab be cd cc df' \
+'s/([a-z])\1/xx/g' \
+'ab be cd xx df'
+
+# Look-around not supported in POSIX
+
+# Multiplicity
+
+checkb '0 or 1' \
+'bb bab baab baa?b baaab' \
+'s/baa\?b/x/g' \
+'bb x x baa?b baaab'
+checke '0 or 1' \
+'bb bab baab baa?b baaab' \
+'s/baa?b/x/g' \
+'bb x x baa?b baaab'
+
+checkb '0 or 1 (negative)' \
+'bb bab baab baa?b baaab' \
+'s/baa?b/x/g' \
+'bb bab baab x baaab'
+checke '0 or 1 (negative)' \
+'bb bab baab baa?b baaab' \
+'s/baa\?b/x/g' \
+'bb bab baab x baaab'
+
+checkb '1 or more' \
+'bb bab baab ba+b baaab' \
+'s/ba\+b/x/g' \
+'bb x x ba+b x'
+checke '1 or more' \
+'bb bab baab ba+b baaab' \
+'s/ba+b/x/g' \
+'bb x x ba+b x'
+
+checkb '1 or more (negative)' \
+'bb bab baab ba+b baaab' \
+'s/ba+b/x/g' \
+'bb bab baab x baaab'
+checke '1 or more (negative)' \
+'bb bab baab ba+b baaab' \
+'s/ba\+b/x/g' \
+'bb bab baab x baaab'
+
+checkb 'Specific number (1)' \
+'bb bab baab baaab baaaab' \
+'s/ba\{2\}b/x/g' \
+'bb bab x baaab baaaab'
+checke 'Specific number (1)' \
+'bb bab baab baaab baaaab' \
+'s/ba{2}b/x/g' \
+'bb bab x baaab baaaab'
+
+checkb 'Specific number (closed)' \
+'bb bab baab baaab baaaab' \
+'s/ba\{1,3\}b/x/g' \
+'bb x x x baaaab'
+checke 'Specific number (closed)' \
+'bb bab baab baaab baaaab' \
+'s/ba{1,3}b/x/g' \
+'bb x x x baaaab'
+
+checkb 'Specific number (open left)' \
+'bb bab baaab baaaab' \
+'s/ba\{,3\}b/x/g' \
+'x x x baaaab'
+checke 'Specific number (open left)' \
+'bb bab baaab baaaab' \
+'s/ba{,3}b/x/g' \
+'x x x baaaab'
+
+checkb 'Specific number (open right)' \
+'bb bab baaab baaaab' \
+'s/ba\{2,\}b/x/g' \
+'bb bab x x'
+checke 'Specific number (open right)' \
+'bb bab baaab baaaab' \
+'s/ba{2,}b/x/g' \
+'bb bab x x'