8,236
edits
Difference between revisions of "User talk:Kelson"
Jump to navigation
Jump to search
→Indexing problem
(→Indexing problem: new section) |
|||
| Line 144: | Line 144: | ||
Hi, please help me. I did download the "wikipedia_en_all_nopic_01_2012.zim" (which is not Pre-indexed), it asked me for indexing before, so I chose YES. At first it took a time about 15 minutes for indexing each percent, but now after 30 hours, it just progressed 69% and make a percent in about one hour (or event more). I think it won't complete indexing with this speed! Are there anyways to index this ZIM file without the Kiwix directly? Thanks [[User:Mariusz|Mariusz]] 12:32, 4 November 2012 (CET) | Hi, please help me. I did download the "wikipedia_en_all_nopic_01_2012.zim" (which is not Pre-indexed), it asked me for indexing before, so I chose YES. At first it took a time about 15 minutes for indexing each percent, but now after 30 hours, it just progressed 69% and make a percent in about one hour (or event more). I think it won't complete indexing with this speed! Are there anyways to index this ZIM file without the Kiwix directly? Thanks [[User:Mariusz|Mariusz]] 12:32, 4 November 2012 (CET) | ||
: Hi Mariusz. Yes, the indexing is really slow, but we can quickly index something like 5 millions of HTML documents. There is also a command tool called kiwix-index to do the job on the command line, but this will be exactly as slow and you will have to modify afterwards things manually (in library.xml) that Kiwix understands where is the new idx path... So, if that too long, I propose you simply to work without the fulltext index. What I can say in addition is: (1) we have fixed the problem that ZIP files do not have the index (like announced) (2) We will allow soon on Kiwix do load a idx file/directory from a custom path. Thank your for your feedback, feel free to report again any problem/whish. Regards [[User:Kelson|Kelson]] 12:43, 4 November 2012 (CET) | |||