schema

schema copied to clipboard

Compute an ngram field for all street names

3

This PR adds one new field called `address_parts.street.ngram` It takes advantage of the `fields` mapping to generate ngrams for street names. The ngram field is tokenized using a new tokenizers...

missinglink

Catch index already exists exception

7

In our deployment, the schema container fails on the create_index job if the schema already exists in elasticsearch. This patch resolves that issue and breaks the cycle of k8s constantly...

gsoyka

This will create a new collection of synonyms specifically for venues, so they can be normalised using OSM tags, and then all variations can be mapped to the normalised values...

NickStallman

WIP, I found this on my HDD and didn't want to lose it. It's an experimental attempt at transliterating Hebrew based off a bug report. If I recall correctly it...

missinglink

expand umlaut, eg. 'ö' -> 'oe'

2

it's common for users speaking germanic languages to expand `ö` to `oe`, especially when stuck using an English keyboard. an example of this is `Löhningen` which we can currently surface...

missinglink

for some reason code in this repo doesn't seem to conform to our usual jslint config. just noticed this code: ``` if( isForced() ) drop(); else prompt( drop, fail );...

missinglink

Synonyms/Suffixes

1

Review our synonyms and suffixes compared to [Nominatim's list](https://github.com/twain47/Nominatim/blob/80df4d3b560f5b1fd550dcf8cdc09a992b69fee0/module/tokenstringreplacements.inc) & [USPS List of Official Abbreviations](https://web.archive.org/web/20140701000000*/https://www.usps.com/send/official-abbreviations.htm) related to: https://github.com/pelias/schema/issues/62

missinglink