261474 - Netscape bookmark file containing javascript can't be imported.

Status: RESOLVED FIXED

(Camino Graveyard :: Bookmarks, defect)

Description

[Hiro]

When Bookmarklet of URL is Import, following bookmark is not import normally.


  -- <DT><A HREF="javascript:var bsetPartialSource = ...
  -- <DT><A HREF="javascript:var bsetLinkTag = ...

In Windows branch/trunk Firefox, it has Imported satisfactory.

Mac OS X 10.3.5
2004092500 (v0.8+)

Comment 1

[Jasper]

Reporter could you please provide some more information. What application
created the hml bookmark file?

I tested and I get a dialog informing me that the file is not supported by the
Camino importer. Note that any how I get a Imported folder in the Boomark
manager. I believe that this is incorrect behaviour.

Comment 2

[Hiro]

(In reply to comment #1)
> Reporter could you please provide some more information. What application
> created the hml bookmark file?

Probably this Bookmarklet is created by the manual.
So,since this Bookmarklet is not what I created, I do not understand about a
more detailed thing...

Comment 3

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

I just tried this again and didn't get a warning dialogue.  In fact, the entire file imported except for one bookmark: リンクタグ作成 (the second bookmark after the first menu spacer, aka the bsetLinkTag item in comment 0).

The previous bookmark (選択部分のソース aka bsetPartialSource) had <> and then リンクタグ作成 appended to its tile when imported (since リンクタグ作成 wasn't imported at all).

Comment 4

[Chris Lawson (gone)]

Is it possible we're crapping out after bsetPartialSource because that bookmarklet is so friggin' long?

Also, we're deliberately ignoring character sets in bookmark files. Does/should this matter?

I'm going to try this myself with the patch for bug 309008 applied and see if that makes any difference.

cl

Comment 5

[Chris Lawson (gone)]

When I import this file with the latest nightly, here's what happens:

1) I get an empty folder, titled "Imported Bookmarks".
2) I get a second folder, titled "べんりセット", containing one bookmark. This bookmark is titled "ID/NAME を表示" and contains the bsetShowIDandNAME Javascript function in its location field.

cl

Comment 6

[Chris Lawson (gone)]

I get the same results as in comment 5 when I import with the patch for bug 309008 applied, so I don't think that has any bearing on this either way.

cl

Comment 7

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Why are our results so different?  I got all but one bookmark with 2005102704 (v1.0a1+) and Chris got only one bookmark, the first one, with the same(?) build.  I used the "Select a file..." option. 

(According to Jesse Ruderman, bookmarks can have at least 20,000 characters: http://www.squarefree.com/bookmarklets/limits.html)

Comment 8

[Samuel Sidler (old account; do not CC)]

Netscape?... If this gets fixed before 1.0, that's great. But it's not going to block it, so I'm taking it off the list.

Comment 9

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Pushing out until someone can figure out what's going wrong here and why Chris and I get such different results.

(I wonder if some bit of OS code is used in the parsing and it changed between 10.3 and 10.4???)

Comment 10

[Samuel Sidler (old account; do not CC)]

Per Mano, this is probably the Netscape Bookmarks File Format, which is used by FF and Mozilla as well.

If that's true, this should get fixed for 1.0. Retargeting back to 1.0.

Comment 11

[Samuel Sidler (old account; do not CC)]

-> 1.1

Comment 12

[Samuel Sidler (old account; do not CC)]

Attachment: testcase

This is simply the testcase from the URL field so we don't lose it.

Comment 13

[Stuart Morgan]

The parser is a nightmare; I wouldn't be at all surprised if the use of '<' and '>' in the JS is (probably among other things) what's killing us. I don't see us rewriting it for 1.6 (although it would be worth someone looking into whether we could just use a core or system HTML parser for this, rather than trying to keep maintaining the parser).

Comment 14

[Stuart Morgan]

This is fixed by my new parser in bug 439024.

Comment 15

[Stuart Morgan]

Fixed by bug 439024.