[dnikola]

4 years, 12 months ago #14293

Hi Partice,

I am currently indexing (indexed) about 4000 custom post type posts in solr.

I this is settings a have setup in 2.2 https://prntscr.com/p9vdhl
All text that needs to be searched is inside this field transkript

so my questions’s are

1. Did i setup this correctly?

2. If i search for same word but on two different letter type – for example Patrice and Патрице will it return same results for both search ? if no, how i can manage when search for one word to return second one.

3. If I search for Patri for example can it return everything that will match after searched term Patricing Patrice Patria Patriamo end etc… Is it enough just to check this option https://prntscr.com/p9vgzf

Thanks!

[wpsolr]

4 years, 12 months ago #14294

1. You indexed your field in 2.2 settings. But you need to search in your field, and for that use the 2.2 option “Index custom fields and categories” or add your field as a boost in 2.3

2. I don’t understand what you intend to do. Automatic translation during search?

3. Yes, partial search should work fine.

[dnikola]

4 years, 12 months ago #14299

1. ok that is something that i already using on page 2.2 https://prntscr.com/p9vnon but on page 2.3 I am not sure what are 2.3 for (do you have some examples to point me how to use this 🙂 )

2. In my language we have two type of letters Latin and Cyrillic. Some text indexed are Latin, and some are Cyrillic. So if some users are using Latin keyboard will they get results from text indexed with Cyrillic?

3. Ok but i thinks that i have more results to see but search is not finding them… How i can debug this …

[wpsolr]

4 years, 12 months ago #14301

1. With 2.2 option, it’s fine

2. What language are we talking about?

3. Try using the Solr admin with search query like: text:Patri*

[dnikola]

4 years, 12 months ago #14304

1. Ok

2. Serbian

3. Ok i will try, if don’t manage to find way how to do that i will ask 🙂

[wpsolr]

4 years, 12 months ago #14305

2. Does this speak to you?
https://lucene.apache.org/solr/guide/6_6/language-analysis.html#LanguageAnalysis-Serbian

[dnikola]

4 years, 12 months ago #14307

yes.

Where i should check for a config file?

[wpsolr]

4 years, 12 months ago #14308

Edit the schema.xml file installed by the plugin in your index folders.

The field analyser to change is “text”:

<fieldType name="text" class="solr.TextField" positionIncrementGap="100">
    <analyzer type="index">
        <tokenizer class="solr.WhitespaceTokenizerFactory"/>
        <filter class="solr.WordDelimiterFilterFactory" generateWordParts="1" generateNumberParts="1"
                catenateWords="1" catenateNumbers="1" catenateAll="0" splitOnCaseChange="1"/>
        <filter class="solr.LowerCaseFilterFactory"/>
        <filter class="solr.EnglishPossessiveFilterFactory"/>
        <filter class="solr.PorterStemFilterFactory"/>
    </analyzer>
    <analyzer type="query">
        <tokenizer class="solr.WhitespaceTokenizerFactory"/>
        <!--<filter class="solr.SynonymFilterFactory" synonyms="synonyms.txt" ignoreCase="true" expand="true"/>-->
        <filter class="solr.WordDelimiterFilterFactory" generateWordParts="1" generateNumberParts="1"
                catenateWords="0" catenateNumbers="0" catenateAll="0" splitOnCaseChange="1"/>
        <filter class="solr.LowerCaseFilterFactory"/>
        <filter class="solr.EnglishPossessiveFilterFactory"/>
        <filter class="solr.PorterStemFilterFactory"/>
    </analyzer>
</fieldType>

[dnikola]

4 years, 12 months ago #14315

ok seams that i have already made some changes

but also i made for query

should i remove?

if i leave this for text will it be ok? https://prntscr.com/p9x6e1

thanks!

[wpsolr]

4 years, 12 months ago #14316

You can use the example from the Solr documentation:

<fieldType name="text" class="solr.TextField" positionIncrementGap="100">
 <analyzer>
  <tokenizer class="solr.StandardTokenizerFactory"/>
  <filter class="solr.LowerCaseFilterFactory"/>
  <filter class="solr.SerbianNormalizationFilterFactory" haircut="bald"/>
 </analyzer>
</fieldType>

Then restart Solr, or “reload” the index, and reindex everything in the plugin.

[dnikola]

4 years, 12 months ago #14317

ok i tried bald and regular but in both cases i get same results.

So my question is are serialized meta fields making some kind of problems with search?
I checked option to debug while indexing and no error has been seen.

But when i search for a specific word i know that there are 35 posts in one category containing that word but it gives me only 33 of that.
I know exactly which posts are not shown.

What i can do about that?

Thanks!

[wpsolr]

4 years, 12 months ago #14318

The 2 posts could be not indexed, in the wrong category, or the keywords searched not in the post title/content (or not exactly).

Try and verify manually with Solr queries in the Solr admin.

[dnikola]

4 years, 12 months ago #14319

I know that post should be in results, but it is not there, it is not up to categories, i dont have post text,search depends on post meta field with serialized data, but other works fine…

Should some error be trown on indexing if post could not be indexed?

[wpsolr]

4 years, 12 months ago #14321

Try to locate your 2 missing posts in your Solr admin search by their ID (query is: PID:xxxx), and check their fields content.

[dnikola]

4 years, 12 months ago #14354

Ok,

i have done some research and what i have noticed that this filed for this posts are partial indexed.

What i have noticed is that it breaks because of >> characters.

for example there is string like

>> TEXT TEXT << text between >> MORE TEXT <<

so basically this text between is skipped, like it is ignoring text between << >>