summaryrefslogtreecommitdiff
path: root/bucky
diff options
context:
space:
mode:
Diffstat (limited to 'bucky')
-rw-r--r--bucky/app/router.js8
-rw-r--r--bucky/search/bdb.js69
-rw-r--r--bucky/search/middleware.js51
-rw-r--r--bucky/search/search.js20
-rw-r--r--bucky/search/snippet.js47
5 files changed, 129 insertions, 66 deletions
diff --git a/bucky/app/router.js b/bucky/app/router.js
index 8104bd5..2fd8d83 100644
--- a/bucky/app/router.js
+++ b/bucky/app/router.js
@@ -91,7 +91,13 @@ module.exports = function(app){
function(req, res){
// delete a comment
})
-
+
+ app.get("/search/",
+ middleware.ensureAuthenticated,
+ function(req, res){
+ res.render("pages/search", {title: "search" })
+ }
+ )
app.get("/api/search",
middleware.ensureAuthenticated,
search.search,
diff --git a/bucky/search/bdb.js b/bucky/search/bdb.js
index ba0124d..a7ced4a 100644
--- a/bucky/search/bdb.js
+++ b/bucky/search/bdb.js
@@ -1,38 +1,55 @@
var bdb_lib = require('berkeleydb')
var dbenv = new bdb_lib.DbEnv();
var bdb_status = dbenv.open('./search/db/env')
-console.log('open /search/db:', bdb_status)
-
-var db
-
-function exitHandler(options, err) {
- db.close()
- // if (options.cleanup) console.log('clean');
- if (err) console.log(err.stack);
- if (options.exit) process.exit();
+if (bdb_status) {
+ console.log('open dbenv failed:', bdb_status)
+ process.exit()
}
-// do something when app is closing
-process.on('exit', exitHandler.bind(null, {cleanup: true}));
+function db(fn){
+ var db
+ fn = "./" + fn + ".db"
-// catches ctrl+c event
-process.on('SIGINT', exitHandler.bind(null, {exit: true}));
+ function exitHandler(options, err) {
+ db.close()
+ // if (options.cleanup) console.log('clean');
+ if (err) console.log(err.stack);
+ if (options.exit) process.exit();
+ }
-// catches "kill pid" (for example: nodemon restart)
-process.on('SIGUSR1', exitHandler.bind(null, {exit: true}));
-process.on('SIGUSR2', exitHandler.bind(null, {exit: true}));
+ // do something when app is closing
+ process.on('exit', exitHandler.bind(null, {cleanup: true}));
-//catches uncaught exceptions
-process.on('uncaughtException', exitHandler.bind(null, {exit:true}));
+ // catches ctrl+c event
+ process.on('SIGINT', exitHandler.bind(null, {exit: true}));
-function open(){
- if (db) db.close()
- var _db = new bdb_lib.Db(dbenv);
- var bdb_status = _db.open('./search.db')
- console.log('open ./search.db:', bdb_status)
- db = _db
-}
+ // catches "kill pid" (for example: nodemon restart)
+ process.on('SIGUSR1', exitHandler.bind(null, {exit: true}));
+ process.on('SIGUSR2', exitHandler.bind(null, {exit: true}));
-open()
+ //catches uncaught exceptions
+ process.on('uncaughtException', exitHandler.bind(null, {exit:true}));
+ function open(fn){
+ if (db) db.close()
+ var _db = new bdb_lib.Db(dbenv);
+ var bdb_status = _db.open(fn)
+ if (bdb_status) {
+ console.log('open ' + fn + ' failed:', bdb_status)
+ process.exit()
+ }
+ db = _db
+ }
+
+ open(fn)
+
+ return {
+ put: function(term, serialized){
+ db.put(term, serialized)
+ },
+ get: function(term){
+ return db.get(term)
+ },
+ }
+}
module.exports = db
diff --git a/bucky/search/middleware.js b/bucky/search/middleware.js
index 39d7a71..32e3321 100644
--- a/bucky/search/middleware.js
+++ b/bucky/search/middleware.js
@@ -6,16 +6,26 @@ module.exports = {
search: function (req, res, next) {
res.search = search.search(req.query.query, req.query.start, req.query.limit)
- console.log(res.search)
+ if (! res.search) {
+ res.sendStatus(400)
+ return
+ }
next()
},
getThreads: function (req, res, next){
var thread_ids = res.search.thread_ids;
if (! thread_ids || ! thread_ids.length) {
+ res.search.threads = []
return next()
}
db.getThreadsById(thread_ids).then(function(threads){
+ threads.forEach((thread) => {
+ var flag_id = thread.get('flagged')
+ if (flag_id) {
+ res.search.file_ids.push(flag_id)
+ }
+ })
res.search.threads = threads
next()
})
@@ -24,11 +34,14 @@ module.exports = {
getComments: function (req, res, next){
var comment_ids = res.search.comment_ids;
if (! comment_ids || ! comment_ids.length) {
+ res.search.comments = []
return next()
}
db.getCommentsById(comment_ids).then(function(comments){
+ var terms = res.search.meta.terms
comments.forEach(function(comment){
- comment.set('comment', comment.get('comment').toString())
+ const snip = snippet(comment.get('comment').toString(), terms)
+ comment.set('comment', snip)
})
res.search.comments = comments
next()
@@ -38,6 +51,7 @@ module.exports = {
getFiles: function (req, res, next){
var file_ids = res.search.file_ids
if (! file_ids || ! file_ids.length) {
+ res.search.files = []
return next()
}
db.getFilesById(file_ids).then(function(files){
@@ -52,7 +66,38 @@ module.exports = {
},
success: function(req, res, next){
- res.send(res.search)
+ var terms = res.search.meta.terms
+ var threads = {}, comments = {}, files = {}
+ res.search.threads.forEach((t) => { threads[t.id] = t })
+ res.search.comments.forEach((t) => { comments[t.id] = t })
+ res.search.files.forEach((t) => { files[t.id] = t })
+ var results = res.search.results.map((r) => {
+ var m = {}
+ m.thread = threads[r.thread]
+ m.comment = comments[r.comment]
+ m.file = files[r.file]
+ m.count = r.count
+ m.strength = r.strength
+ if (m.thread) {
+ var flagged = m.thread.get('flagged')
+ if (flagged) {
+ m.thread.set('flagged', files[flagged])
+ }
+ var allowed = m.thread.get('allowed')
+ if (allowed) {
+ m.thread.set('allowed', allowed.toString().split(" "))
+ }
+ var display = m.thread.get('display')
+ if (display) {
+ m.thread.set('display', display.toString().split(" "))
+ }
+ }
+ return m
+ })
+ res.send({
+ meta: res.search.meta,
+ results: results,
+ })
},
}
diff --git a/bucky/search/search.js b/bucky/search/search.js
index a28d49c..4818ef9 100644
--- a/bucky/search/search.js
+++ b/bucky/search/search.js
@@ -1,5 +1,5 @@
var db = require('../db')
-var bdb = require('./bdb')
+var bdb = require('./bdb')('search')
var STOPWORDS = require('./stopwords')
var wordRegexp = new RegExp("(\W+)");
@@ -28,7 +28,7 @@ function find_term(term) {
strength: parseInt(partz[3]) || 1,
}
})
- console.log(matches)
+ // console.log(matches)
return matches
}
@@ -69,7 +69,7 @@ function search (query, start, limit) {
if (i < start) return false
if (to_display-- === 0) return true
results.push(match)
- console.log(match)
+ // console.log(match)
thread_ids.push(match.thread)
if (match.comment) comment_ids.push(match.comment)
if (match.file) file_ids.push(match.file)
@@ -77,16 +77,18 @@ function search (query, start, limit) {
})
return {
- query: query,
- start: start,
- next: start + limit,
- limit: limit,
- total: total,
+ meta: {
+ query: query,
+ terms: terms,
+ start: start,
+ next: start + limit,
+ limit: limit,
+ total: total,
+ },
results: results,
thread_ids: thread_ids,
comment_ids: comment_ids,
file_ids: file_ids,
- terms: terms,
};
}
diff --git a/bucky/search/snippet.js b/bucky/search/snippet.js
index cd0657f..17988d2 100644
--- a/bucky/search/snippet.js
+++ b/bucky/search/snippet.js
@@ -1,19 +1,11 @@
var util = require('../util/util')
var STOPWORDS = require('./stopwords')
-function bold_snippet(s, terms) {
- return bold_terms(snippet(s, terms), terms)
-}
-function bold_terms (s, terms) {
- s = util.sanitize(s)
- terms.forEach( (term) => {
- s.replace(new RegExp("\b" + term + "\b", "i"), "<b>" + term + "</b>")
- })
-}
function snippet(s, terms) {
s = util.sanitize(s)
- var term_re = new RegExp("\b(" + terms.join("|") + ")\b", "i")
- var words = s.split(/\s+/)
+ var term_set = new Set(terms)
+
+ var words = s.split(/[^a-zA-Z0-9]+/)
var snippet = "";
// deduper for matching @words indexes, so we don't add a word twice
@@ -26,27 +18,30 @@ function snippet(s, terms) {
var aggr = 0;
// amount of context to show, in number of words surrounding a match
- var $pad = 4;
+ var pad = 10;
// loop over each of the words in the string
- words.some((word, i) => {
+ var word
+ for (var i = 0, len = words.length; i < len; i++) {
+ word = words[i]
+
// if the word matches...
- if (term_re.match(word) && ! STOPWORDS.has(word.toLowerCase())) {
+ if (term_set.has(word.toLowerCase()) && ! STOPWORDS.has(word.toLowerCase())) {
// if we aren't already aggregating, add an ellipsis
- if (! $aggr) {
+ if (! aggr) {
words_matched.push("...")
}
-
+
// look backward $pad words
var idx;
- for (var j = -pad; j < 1; j++) {
+ INNER: for (var j = -pad; j < 1; j++) {
// create a new index from the offset
idx = i + j;
// is this a valid index? has it already been encountered?
- if (idx < 0) continue;
- if (idx > words.length) continue;
- if (index_matches[idx]) continue;
+ if (idx < 0) continue INNER;
+ if (idx > words.length) continue INNER;
+ if (index_matches[idx]) continue INNER;
// checks out, save this word
words_matched.push(words[idx])
@@ -69,20 +64,18 @@ function snippet(s, terms) {
// one less word to aggregate
aggr--;
}
-
+
// keep snippets to a modest length
- return words_matched.length > 30;
- })
+ if (words_matched.length > 30) break
+ }
// add a trailing ellipsis
words_matched.push("...")
// create the snippet from the saved context words
snippet = words_matched.join(" ")
-
+
return snippet
}
-module.exports = {
- bold_snippet, bold_terms, snippet,
-} \ No newline at end of file
+module.exports = snippet \ No newline at end of file
generated by cgit v1.2.3-70-g09d2 (git 2.51.0) at 2025-12-10 17:21:50 +0000