My company is looking at advanced search and reporting solutions, and are considering (among other options) creating something akin to JIRA's JQL for maximum fl开发者_C百科exibility.
My googling leads me to believe Atlassian built JQL from scratch, at least as a language with syntax and a parser, but I thought I'd try SO before concluding. Anyone know, at a high level, how they did it? Was there one or more Open Source project they based it on?
(Kudos to Atlassian either way - JQL is gorgeous!)
I think they did it from scratch. The underlying architecture is crisp but quite complex. It took me a good few hours to get it, just reading the source and minimal user docs.
~Matt
Atlassian built JQL on top of Apache Lucene. You might want to take a look at Elasticsearch or Solr, which are open source alternatives, also built on Lucene.
I have been using Jira for a year and I notice "Apache Lucene" on the the directory, and before this I got a job wherein I was force to learn apache solr. So in conclusion, Jira is using Apache Lucene as a searching library which is also used was being used in Solr.
for more info read this: http://www.lucenetutorial.com/lucene-vs-solr.html
精彩评论