elasticsearch regexp don't work

Question

I need to make a regexp on elasticsearch to filtre some data. The field I filter on is the name of person. The data are not always well formatted (sometimes, there is no first name, sometimes, the family name is followed by a period or a comma or 'comma+first name' or 'point+first name'....).

For example, using "bouchard" I get the following matches:

 "bouchard", "bouchard, m.", "bouchard, j.", "bouchard j.p.", "bouchard. j.p."

I need also to exclude name who begin with same prefixe like "bouchardat".

I tried many regexps and finally found that an exclusion may yield better results:

   "query" :  { "regexp" : {
                    "RECORDEDBY" : "bouchard([^a-z].*)"
    }}

This doesn't work because it returns "bouchard, m.", "bouchard, j.", "bouchard j.p." but not "bouchard. j.p." and not "bouchard".

I try some regexps with + and .* but they don't work.

( "bouchard([^a-z].*.*)" "bouchard([^a-z]*+.*)")

To make it clear, I want to allow:

bouchard
bouchard, m.
bouchard, j.
bouchard j.p.
bouchard. j.p.

I want to exclude

bouchardat

Any advice is welcome.

Could you please be more specific? What entries do you allow, and which ones do you want to disallow? The documentation says that Elasticsearch regexes are always anchored, so "RECORDEDBY" : "bouchard" will only allow bouchard, and "RECORDEDBY" : "bouchard.+" should allow any values starting with bouchard. — Wiktor Stribiżew
– Wiktor Stribiżew, Commented Mar 30, 2015 at 10:14
Sorry. I want to exclude "bouchardat" and to allow : "bouchard", "bouchard, m.", "bouchard, j.", "bouchard j.p.","bouchard. j.p." and all the entries with the same name followed by a combination of space/point/coma and any word. — AlainIb
– AlainIb, Commented Mar 30, 2015 at 10:55
bouchard[^a-zA-Z]* return only "bouchard" but not "bouchard, m.", "bouchard, j.", "bouchard j.p.","bouchard. j.p." — AlainIb
– AlainIb, Commented Mar 30, 2015 at 10:58
"bouchard[^a-zA-Z]*.*" return "bouchard","bouchard, m.","bouchard, j.","bouchard j.p.","bouchardat" . it miss "bouchard. j.p." and allow ""bouchardat" — AlainIb
– AlainIb, Commented Mar 30, 2015 at 11:01

frhd · Accepted Answer · 2015-03-30 12:37:54Z

1

In this case, you could use a conditional operator to exclude every [a-z] suffix if no special character like '', '.', or ',' follows the word you are looking for:

((bouchard)+?([ .,]+)[ ,.a-zA-Z]*)|(bouchard[^a-zA-Z]?)

This regexp returns for the condition (there has to be [ .,]+):

bouchard
bouchard, m.
bouchard, j.
bouchard j.p.
bouchard. j.p.

and ignores the stuff after the pipe | where no [ .,]+ applies:

bouchardat

Regex101

edited Mar 30, 2015 at 12:37

answered Mar 30, 2015 at 11:26

frhd

10.4k5 gold badges27 silver badges43 bronze badges

Sign up to request clarification or add additional context in comments.

8 Comments

AlainIb Over a year ago

thanks for the explained solution. unfortunaly it don't return "bouchard". the caractere comma, space and point didn't need to be escaped ?

frhd Over a year ago

Of course, the ? was missing at the end. Updated and added reges101 link.

frhd Over a year ago

And of course, you want to capture the whole group, so updated once more.

AlainIb Over a year ago

thank you very much. thank for the Regex101 site too.

Tensibai Over a year ago

(bouchard+[^a-zA-Z]?) this one will match bouchardddddd (infite number of d) I don't see the reason of this + after d at all

|

Collectives™ on Stack Overflow

elasticsearch regexp don't work

1 Answer 1

8 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

8 Comments

Your Answer

Sign up or log in

Post as a guest

Related