summaryrefslogtreecommitdiff
path: root/doc/bugs
diff options
context:
space:
mode:
Diffstat (limited to 'doc/bugs')
-rw-r--r--doc/bugs/attachment:_escaping_underscores_in_filename__63__.mdwn3
-rw-r--r--doc/bugs/markdown_bug:_email_escaping_and_plus_addresses.mdwn36
-rw-r--r--doc/bugs/ssl_certificates_not_checked_with_openid.mdwn16
-rw-r--r--doc/bugs/toggle_fails_on_Safari.mdwn58
4 files changed, 113 insertions, 0 deletions
diff --git a/doc/bugs/attachment:_escaping_underscores_in_filename__63__.mdwn b/doc/bugs/attachment:_escaping_underscores_in_filename__63__.mdwn
index d7d101700..4ce4ac5ee 100644
--- a/doc/bugs/attachment:_escaping_underscores_in_filename__63__.mdwn
+++ b/doc/bugs/attachment:_escaping_underscores_in_filename__63__.mdwn
@@ -17,3 +17,6 @@ Is it a bug or security feature? --[[Paweł|ptecza]]
>> (`myisam__95__vs__95__ndb` instead of `myisam_vs_ndb`). --[[Paweł|ptecza]]
> [[done]], uses `linkpage` now.
+
+>> It's seems that now Ikiwiki doesn't escape the filenames with underscore(s).
+>> Thank you very much for the fast fix! --[[Paweł|ptecza]]
diff --git a/doc/bugs/markdown_bug:_email_escaping_and_plus_addresses.mdwn b/doc/bugs/markdown_bug:_email_escaping_and_plus_addresses.mdwn
new file mode 100644
index 000000000..8efd6da57
--- /dev/null
+++ b/doc/bugs/markdown_bug:_email_escaping_and_plus_addresses.mdwn
@@ -0,0 +1,36 @@
+compare:
+
+ * <jon+markdownbug@example.org>
+ * <jon.markdownbug@example.org>
+
+* <jon+markdownbug@example.org>
+* <jon.markdownbug@example.org>
+
+It seems putting a '+' in there throws it. Maybe it's a markdown bug, or maybe the obfuscation markdown applies to email-links is being caught by the HTML sanitizer.
+
+ -- [[JonDowland]]
+
+> It's a markdown bug. For some reason, markdown doesn't recognize the email with a '+' as an email:
+>
+> $ echo '<a+b@c.org>' | markdown
+> <p><a+b@c.org></p>
+>
+> htmlscrubber then (rightly) removes this unknown tag.
+>
+
+>> Filed [in CPAN](http://rt.cpan.org/Ticket/Display.html?id=37909) --[[Joey]] [[tag done]]
+
+> But I've noticed some other Text::Markdown bugs that, even with htmlscrubber, produce
+> [ill-formed (X)HTML](http://validator.w3.org/check?uri=http%3A%2F%2Fikiwiki.info%2Fbugs%2Fmarkdown_bug%3A_email_escaping_and_plus_addresses%2F).
+> (View the markdown source of this page.)
+>
+> --Gabriel
+
+>> The htmlscrubber does not attempt to produce valid html from invalid. It
+>> attempts to prevent exploits in html. The tidy plugin can force html to
+>> valid. --[[Joey]]
+
+<tt>
+
+-
+>
diff --git a/doc/bugs/ssl_certificates_not_checked_with_openid.mdwn b/doc/bugs/ssl_certificates_not_checked_with_openid.mdwn
index cb4c706f0..e3bd56cfd 100644
--- a/doc/bugs/ssl_certificates_not_checked_with_openid.mdwn
+++ b/doc/bugs/ssl_certificates_not_checked_with_openid.mdwn
@@ -12,6 +12,10 @@ For now, I want to try and resolve the issues with net\_ssl\_test, and run more
> ikiwiki) performing any sanity checking of the openid server. All the
> security authentication goes on between your web browser and the openid
> server. This may involve ssl, or not.
+>
+>> Note that I'm not an openid expert, and the above may need to be taken
+>> with a grain of salt. I also can make no general statements about openid
+>> being secure. ;-) --[[Joey]]
>
> For example, my openid is "http://joey.kitenet.net/". If I log in with
> this openid, ikiwiki connects to that http url to determine what openid
@@ -34,3 +38,15 @@ For now, I want to try and resolve the issues with net\_ssl\_test, and run more
>> for use by ikiwiki and the rest is simple.
>> -- Brian May
+
+>>> I guess that the place to add SSL cert checking would be in either
+>>> [[cpan LWPx::ParanoidAgent]] or [[cpan Net::OpenID::Consumer]]. Adding
+>>> it to ikiwiki itself, which is just a user of those libraries, doesn't
+>>> seem right.
+>>>
+>>> It's not particularly clear to me how a SSL cert can usefully be
+>>> checked at this level, where there is no way to do anything but
+>>> succeed, or fail; and where the extent of the check that can be done is
+>>> that the SSL cert is issued by a trusted party and matches the domain name
+>>> of the site being connected to. I also don't personally think that SSL
+>>> certs are the right fix for DNS poisoning issues. --[[Joey]]
diff --git a/doc/bugs/toggle_fails_on_Safari.mdwn b/doc/bugs/toggle_fails_on_Safari.mdwn
new file mode 100644
index 000000000..25f62e088
--- /dev/null
+++ b/doc/bugs/toggle_fails_on_Safari.mdwn
@@ -0,0 +1,58 @@
+The [[plugins/toggle]] plugin has no effect when viewed on the Safari web browser.
+
+All toggles appear open all the time.
+
+I don't know if this is true for other webkit browsers (the new Konqueror, the iPhone, etc).
+I'm currently testing in the Safari nightly builds, but I've seen the bug in the current release
+of Safari too.
+
+Looking at the Safari Web Inspector, it believes there is a parse error on line 47 of the
+[[news]] page. This is the definition of the getElementsByClass(class) function.
+
+ 45 }
+ 46
+ 47 function getElementsByClass(class) {
+ SyntaxError: Parse error
+ 48 var ret = new Array();
+
+> Reproduced in epiphany-webkit on debian.
+>
+> Also noticed something interesting when I opened the page in vim. It
+> highlighted the "class" like a type definition, not a variable. Sure
+> enough, replacing with "c" fixed it.
+>
+> I wonder if webkit is actually in the right here, and using a reseved
+> word like, presumably, "class" as a variable name is not legal. As I try
+> to ignore javascript as much as possible, I can't say. [[done]] --[[Joey]]
+
+>> I also started having a look at this. I found the same issue with the
+>> the variable 'class'. I'm not a javascript guru so I looked on the web
+>> at other implementations of getElementsByClass() and noticed some
+>> things that we might use. I took a bunch of different ideas and came
+>> up with this:
+
+ function getElementsByClass(cls, node, tag) {
+ if (document.getElementsByClass)
+ return document.getElementsByClass(cls, node, tag);
+ if (! node) node = document;
+ if (! tag) tag = '*';
+ var ret = new Array();
+ var pattern = new RegExp("(^|\\s)"+cls+"(\\s|$)");
+ var els = node.getElementsByTagName(tag);
+ for (i = 0; i < els.length; i++) {
+ if ( pattern.test(els[i].className) ) {
+ ret.push(els[i]);
+ }
+ }
+ return ret;
+ }
+
+>> Most of the changes are minor, except that this one will use the
+>> built in function if it is available. That is likely to be significantly
+>> faster. Adding the extra parameters doesn't cause a problem --
+>> they're filled in with useful defaults.
+
+>> I don't know if it is worth making this change, but it is there if you want it.
+
+>>> Well, it seems to work. Although god only knows about IE. Suppose I
+>>> might as well.. --[[Joey]]