Incorrect placement of line breaks inside HTML <pre> tags

[vincent.privat@…]

Hello, We're using Trac 1.2.2.

When entering on a Wiki page:

# Running a jar
java [Java options] -jar josm-tested.jar [Program arguments]

# Launch a Web Start
javaws [run-options] -J[Java option] josm.jnlp

The resulting HTML code is:

<div class="wiki-code"><div class="code"><pre><span class="c1"># Running a jar
</span>java <span class="o">[</span>Java options<span class="o">]</span> -jar josm-tested.jar <span class="o">[</span>Program arguments<span class="o">]</span>

<span class="c1"># Launch a Web Start
</span>javaws <span class="o">[</span>run-options<span class="o">]</span> -J<span class="o">[</span>Java option<span class="o">]</span> josm.jnlp
</pre></div></div>

This causes a problem when we try to display this HTML code with Swing (​https://josm.openstreetmap.de/ticket/15998) probably because the line break is inside the <span> instead of being right after it.

I think the correct HTML code should be:

<div class="wiki-code"><div class="code"><pre><span class="c1"># Running a jar</span>
java <span class="o">[</span>Java options<span class="o">]</span> -jar josm-tested.jar <span class="o">[</span>Program arguments<span class="o">]</span>

<span class="c1"># Launch a Web Start</span>
javaws <span class="o">[</span>run-options<span class="o">]</span> -J<span class="o">[</span>Java option<span class="o">]</span> josm.jnlp
</pre></div></div>

[vincent.privat@…]

It seems the behaviour is correct on this Trac instance so it might be a duplicate of another ticket fixed meanwhile.

[anonymous]

Bisecting suggests it was fixed by [15418].

[Jun Omae]

I think that behavior is a html viewer's issue in Swing. Newlines even within span element in pre element shouldn't be ignored.

trac/mimeview/pygments.py

@@ -295 +295 @@
         def _generate():
             for c, text in self._chunk(tokens):
                 if c:
-                    attrs = Attrs([(class_, c)])
-                    yield START, (span, attrs), pos
-                    yield TEXT, text, pos
-                    yield END, span, pos
+                    for idx, t in enumerate(text.split('\n')):
+                        if t:
+                            attrs = Attrs([(class_, c)])
+                            yield START, (span, attrs), pos
+                            yield TEXT, t, pos
+                            yield END, span, pos
+                        if idx > 0:
+                            yield TEXT, '\n', pos
                 else:
                     yield TEXT, text, pos
         return Stream(_generate())