On Fri, Jul 15, 2016 at 2:33 PM, Lyric Wei <[email protected]> wrote:
> I think the fix you mentioned is correct. If I recall it's related to a >> change between Django 1.8 and 1.9, and we made the same fix in Mezzanine >> already, but Drum was not updated presumably. I can submit the fix if you >> really don't want to, but I encourage you to. >> >> > Oh, I am glad to do that.. I don't find source code of Mezzanine in Drum > project, but I got this: > https://github.com/stephenmcd/drum/blob/master/setup.py#L54 . Since you > have made the fix in latest Mezzanine, I can simply change this line, from > "mezzanine >= 4.0.1" to "mezzanine >= 4.1.0", am i right? > That needs to happen, but it's separate from the fix - what I meant was that Drum literally needs the same change that was applied to Mezzanine. I've bumped the version here: https://github.com/stephenmcd/drum/commit/177a202e012ac4c65a5aa2aa289b11008a1c62b7 And made the fix here: https://github.com/stephenmcd/drum/commit/7b327615a93ee4fa04d595b53d20f9d9f34db925 > > That's excellent info, thanks for going into detail there. >> >> Do you think it's possible to programmatically extract tags in those >> languages, using different logic? If so we can make the extract function >> configurable by the developer. >> > > Yes. Search engines like Google need this kind of technology to create > indexes for each words to support different languages. And there are > several existed open source projects to extract tags in those languages. > > A famous project named JieBa(https://github.com/fxsjy/jieba): a chinese > text segmentation project in Python, which use dictionary and statistical > result to extract words, also works for Japanese if we provide suitable > dictionary. > I've made the tag extraction configurable now, see here: https://github.com/stephenmcd/drum/commit/147984d813c54996e0b97f32959fdd3992bf509d > > > On Friday, July 15, 2016 at 7:38:45 AM UTC+8, Stephen McDonald wrote: > >> >> >> On Thu, Jul 14, 2016 at 5:31 PM, Lyric Wei <[email protected]> wrote: >> >>> > If you have a bug fix, please go ahead and submit it in a pull request >>> via Github. >>> >>> I am trying to fix it, not yet. I am not familiar with django. But I >>> though it's a bug in mezzanine.core.models. >>> >> >> I think the fix you mentioned is correct. If I recall it's related to a >> change between Django 1.8 and 1.9, and we made the same fix in Mezzanine >> already, but Drum was not updated presumably. I can submit the fix if you >> really don't want to, but I encourage you to. >> >> >>> >>> > What do you mean? >>> >>> I am requesting a new feature, the tags should support non-character >>> based languages, like Japanese, Arabic, Chinese, etc. >>> >>> the current keywords extract approach doesn't support those languages. >>> >>> Maybe we can allow user to add tag when they add new links, manually >>> adding tags, but simple. >>> >>> or >>> >>> Drum support different keywords extract approaches, to extract keywords >>> from non-character based languages. >>> >>> Btw, words in those non-character based languages are not split by >>> *space* and *symbols*.That's why I say "different keywords extract >>> approaches". for example, >>> >>> In English, we say "This morning, the weather is good", can be split as >>> "This", "morning", "the", "weather", "is", "good" by space and symbols. >>> >>> In Japanese, we say 今朝は天気が良いです, can be split as "今朝", "は", "天気", "が", >>> "良い", "です", according to the grammar, not the space and symbols. >>> >>> In Chinese, we say 今天早上天气不错, can be split as "今天", "早上", "天气", "不错", >>> according to the grammar either, not the space and symbols. >>> >> >> That's excellent info, thanks for going into detail there. >> >> Do you think it's possible to programmatically extract tags in those >> languages, using different logic? If so we can make the extract function >> configurable by the developer. >> >> >>> >>> >>> >>> >>> >>> >>> On Thursday, July 14, 2016 at 3:11:21 PM UTC+8, Stephen McDonald wrote: >>> >>>> >>>> >>>> On Thu, Jul 14, 2016 at 4:54 PM, Lyric Wei <[email protected]> wrote: >>>> >>>>> Thank you for quick response. Now django show me the follow error >>>>> message in /usr/local/lib/python2.7/dist-packages/drum/links/models.py >>>>> >>>>> > self.keywords.add(AssignedKeyword(keyword=keyword)) >>>>> >>>>> > instance isn't saved. Use bulk=False or save the object first. >>>>> >>>>> I check out the code and found how to resolve it. I have to pass >>>>> bulk=False to save method to bypass. >>>>> >>>> >>>> If you have a bug fix, please go ahead and submit it in a pull request >>>> via Github. >>>> >>>> >>>>> >>>>> Oh yes, the more important question. Is it possible to add an existed >>>>> tag manually when user adding a link? >>>>> >>>>> Because I am using drum for international site, a lot of languages, >>>>> like Japanese, Korean, Arabic, Chinese, are not supported by the keyword >>>>> extract method drum is using, which i found it in `save` method. >>>>> >>>> >>>> What do you mean? >>>> >>>> >>>>> >>>>> I though there are two approaches to solve the problem, to support >>>>> different language: >>>>> >>>>> 1. allow visitors who submit the link to add tags either. >>>>> 2. allow drum to support different word extract method, which i >>>>> suppose the user can specified in settings.py. >>>>> >>>>> I thought the first is easy, but 2nd is more powerful. >>>>> >>>>> On Thursday, July 14, 2016 at 9:53:22 AM UTC+8, Stephen McDonald wrote: >>>>>> >>>>>> In Mezzanine they're called "keywords", so you should see that. >>>>>> >>>>>> I've just updated the readme to mention that. >>>>>> >>>>>> On Wed, Jul 13, 2016 at 10:31 PM, Lyric Wei <[email protected]> wrote: >>>>>> >>>>>>> Hi there, >>>>>>> >>>>>>> I am using drum(https://github.com/stephenmcd/drum) as a news >>>>>>> social website. drum is a Reddit / Hacker News clone for Mezzanine. >>>>>>> >>>>>>> Its document mentioned that >>>>>>> >>>>>>> > This means that for auto-tagging to work, the tags must already >>>>>>> exist in the database. You can either add them manually via the admin... >>>>>>> >>>>>>> But I can't find the way to add existed tags in the admin panel. >>>>>>> Could you help me? >>>>>>> >>>>>>> Thank you. >>>>>>> >>>>>>> -- >>>>>>> You received this message because you are subscribed to the Google >>>>>>> Groups "Mezzanine Users" group. >>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>> send an email to [email protected]. >>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> Stephen McDonald >>>>>> http://jupo.org >>>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Mezzanine Users" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>> >>>> >>>> >>>> -- >>>> Stephen McDonald >>>> http://jupo.org >>>> >>> >> >> >> -- >> Stephen McDonald >> http://jupo.org >> > -- > You received this message because you are subscribed to the Google Groups > "Mezzanine Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > For more options, visit https://groups.google.com/d/optout. > -- Stephen McDonald http://jupo.org -- You received this message because you are subscribed to the Google Groups "Mezzanine Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
