annotate src/luan/modules/lucene/Lucene.luan @ 1375:5c3702f60200

cleanup
author Franklin Schmidt <fschmidt@gmail.com>
date Thu, 11 Jul 2019 23:19:11 -0600
parents 709f7498a363
children 4c9691d6288f
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
1353
8d95711f6615 replace java() with require "java"
Franklin Schmidt <fschmidt@gmail.com>
parents: 1348
diff changeset
1 require "java"
693
ca169567ce07 module URIs must now include ".luan"
Franklin Schmidt <fschmidt@gmail.com>
parents: 625
diff changeset
2 local Luan = require "luan:Luan.luan"
503
92c3d22745b8 make _ENV optional
Franklin Schmidt <fschmidt@gmail.com>
parents: 435
diff changeset
3 local error = Luan.error
625
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
4 local ipairs = Luan.ipairs or error()
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
5 local type = Luan.type or error()
796
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
6 local set_metatable = Luan.set_metatable or error()
1365
6617763dfd76 fix logging and lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 1353
diff changeset
7 local Boot = require "luan:Boot.luan"
693
ca169567ce07 module URIs must now include ".luan"
Franklin Schmidt <fschmidt@gmail.com>
parents: 625
diff changeset
8 local Html = require "luan:Html.luan"
707
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
9 local Io = require "luan:Io.luan"
746
293c397e8dee improve zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 730
diff changeset
10 local uri = Io.uri or error()
707
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
11 local String = require "luan:String.luan"
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
12 local matches = String.matches or error()
1368
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
13 local Thread = require "luan:Thread.luan"
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
14 local schedule = Thread.schedule or error()
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
15 local run_for_backup = Thread.run_for_backup or error()
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
16 local Time = require "luan:Time.luan"
749
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
17 local Rpc = require "luan:Rpc.luan"
321
7f7708e8fdd4 remove import statement
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 320
diff changeset
18 local LuceneIndex = require "java:luan.modules.lucene.LuceneIndex"
1344
dc2af9d5463b move queryparser to lib
Franklin Schmidt <fschmidt@gmail.com>
parents: 1343
diff changeset
19 local NumberFieldParser = require "java:luan.lib.queryparser.NumberFieldParser"
dc2af9d5463b move queryparser to lib
Franklin Schmidt <fschmidt@gmail.com>
parents: 1343
diff changeset
20 local SaneQueryParser = require "java:luan.lib.queryparser.SaneQueryParser"
1368
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
21 local Logging = require "luan:logging/Logging.luan"
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
22 local logger = Logging.logger "Lucene"
544
c5a93767cc5c lucene overhaul, untested
Franklin Schmidt <fschmidt@gmail.com>
parents: 542
diff changeset
23
320
fed1893821bf remove global namespace
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 312
diff changeset
24
1088
bae2d0c2576c change module naming convention
Franklin Schmidt <fschmidt@gmail.com>
parents: 796
diff changeset
25 local Lucene = {}
230
4438cb2e04d0 start lucene
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents:
diff changeset
26
1088
bae2d0c2576c change module naming convention
Franklin Schmidt <fschmidt@gmail.com>
parents: 796
diff changeset
27 Lucene.type = {
1345
6f8988830098 unique LuceneIndex per dir
Franklin Schmidt <fschmidt@gmail.com>
parents: 1344
diff changeset
28 english = LuceneIndex.ENGLISH_FIELD_PARSER
6f8988830098 unique LuceneIndex per dir
Franklin Schmidt <fschmidt@gmail.com>
parents: 1344
diff changeset
29 string = LuceneIndex.STRING_FIELD_PARSER
6f8988830098 unique LuceneIndex per dir
Franklin Schmidt <fschmidt@gmail.com>
parents: 1344
diff changeset
30 integer = NumberFieldParser.INT
6f8988830098 unique LuceneIndex per dir
Franklin Schmidt <fschmidt@gmail.com>
parents: 1344
diff changeset
31 long = NumberFieldParser.LONG
6f8988830098 unique LuceneIndex per dir
Franklin Schmidt <fschmidt@gmail.com>
parents: 1344
diff changeset
32 double = NumberFieldParser.DOUBLE
544
c5a93767cc5c lucene overhaul, untested
Franklin Schmidt <fschmidt@gmail.com>
parents: 542
diff changeset
33 }
c5a93767cc5c lucene overhaul, untested
Franklin Schmidt <fschmidt@gmail.com>
parents: 542
diff changeset
34
1088
bae2d0c2576c change module naming convention
Franklin Schmidt <fschmidt@gmail.com>
parents: 796
diff changeset
35 Lucene.literal = SaneQueryParser.literal
547
0be287ab0309 add lucene/Versioning and simplify Lucene fn names
Franklin Schmidt <fschmidt@gmail.com>
parents: 546
diff changeset
36
1088
bae2d0c2576c change module naming convention
Franklin Schmidt <fschmidt@gmail.com>
parents: 796
diff changeset
37 function Lucene.index(index_dir,default_type,default_fields)
1369
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
38 type(index_dir)=="table" or error "index_dir must be table"
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
39 index_dir.to_uri_string and matches(index_dir.to_uri_string(),"^file:") or error "must be file"
303
fdb4bd391c28 add lucene close();
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 300
diff changeset
40 local index = {}
591
790d5de23042 add "strict" param to Io.repr();
Franklin Schmidt <fschmidt@gmail.com>
parents: 547
diff changeset
41 index.dir = index_dir
1369
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
42 local java_index, closer = LuceneIndex.getLuceneIndex(index_dir.java.file,default_type,default_fields)
1129
3234a14bb1f8 minor lucene changes
Franklin Schmidt <fschmidt@gmail.com>
parents: 1088
diff changeset
43 index.java = java_index
1345
6f8988830098 unique LuceneIndex per dir
Franklin Schmidt <fschmidt@gmail.com>
parents: 1344
diff changeset
44 index.closer = closer or error()
796
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
45
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
46 index.indexed_fields = {}
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
47 local mt = {}
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
48 set_metatable(index.indexed_fields,mt)
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
49 function mt.__index(_,key)
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
50 return java_index.getIndexedFieldParser(key)
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
51 end
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
52 function mt.__new_index(_,key,value)
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
53 return java_index.setIndexedFieldParser(key,value)
6b8ea0a9b7c9 remove LuanMeta
Franklin Schmidt <fschmidt@gmail.com>
parents: 775
diff changeset
54 end
618
5e495e4e560b add lucene indexed_only_fields
Franklin Schmidt <fschmidt@gmail.com>
parents: 617
diff changeset
55
303
fdb4bd391c28 add lucene close();
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 300
diff changeset
56 index.to_string = java_index.to_string
707
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
57 index.snapshot = java_index.snapshot
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
58 index.advanced_search = java_index.advanced_search
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
59 index.search_in_transaction = java_index.search_in_transaction
303
fdb4bd391c28 add lucene close();
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 300
diff changeset
60 index.delete_all = java_index.delete_all
547
0be287ab0309 add lucene/Versioning and simplify Lucene fn names
Franklin Schmidt <fschmidt@gmail.com>
parents: 546
diff changeset
61 index.delete = java_index.delete
0be287ab0309 add lucene/Versioning and simplify Lucene fn names
Franklin Schmidt <fschmidt@gmail.com>
parents: 546
diff changeset
62 index.save = java_index.save
546
eaef1005ab87 general lucene cleanup
Franklin Schmidt <fschmidt@gmail.com>
parents: 545
diff changeset
63 index.update_in_transaction = java_index.update_in_transaction
591
790d5de23042 add "strict" param to Io.repr();
Franklin Schmidt <fschmidt@gmail.com>
parents: 547
diff changeset
64 index.ensure_open = java_index.ensure_open
617
e54c1646eed0 add Lucene.next_id()
Franklin Schmidt <fschmidt@gmail.com>
parents: 599
diff changeset
65 index.next_id = java_index.nextId
624
8281a248c47e add lucene highlighter
Franklin Schmidt <fschmidt@gmail.com>
parents: 622
diff changeset
66 index.highlighter = java_index.highlighter
1343
7d9a1f8894b0 lucene change indexed_only_field() to indexed_only_fields()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1342
diff changeset
67 index.indexed_only_fields = java_index.indexed_only_fields
1341
a015a0b5c388 add Html.decode(), Lucene.count_tokens(), lucene boosts, Sql.database.set()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1330
diff changeset
68 index.count_tokens = java_index.count_tokens
1375
5c3702f60200 cleanup
Franklin Schmidt <fschmidt@gmail.com>
parents: 1369
diff changeset
69 index.close = closer.close
709
96a280ca32a2 add Lucene.instances
Franklin Schmidt <fschmidt@gmail.com>
parents: 707
diff changeset
70
1342
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
71 function index.search( query, from, to, options )
770
Franklin Schmidt <fschmidt@gmail.com>
parents: 761
diff changeset
72 from or error "missing 'from' parameter"
Franklin Schmidt <fschmidt@gmail.com>
parents: 761
diff changeset
73 to or error "missing 'to' parameter"
1342
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
74 options = options or {}
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
75 local explain_fld = options.explain
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
76 local score_fld = options.score
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
77 local results = {}
1342
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
78 local function fn(i,doc_fn,score)
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
79 if i >= from then
1342
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
80 local doc
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
81 if explain_fld == nil then
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
82 doc = doc_fn()
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
83 else
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
84 local explanation
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
85 doc, explanation = doc_fn("explain")
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
86 doc[explain_fld] = explanation.toString()
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
87 end
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
88 if score_fld ~= nil then
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
89 doc[score_fld] = score
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
90 end
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
91 results[#results+1] = doc
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
92 end
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
93 end
1342
60599adc27b8 add lucene search options
Franklin Schmidt <fschmidt@gmail.com>
parents: 1341
diff changeset
94 local total_hits = index.advanced_search(query,fn,to,options.sort)
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
95 return results, total_hits
257
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
96 end
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
97
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
98 function index.get_document(query)
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
99 local doc
1348
af478a1bd23d bug fix
Franklin Schmidt <fschmidt@gmail.com>
parents: 1345
diff changeset
100 local function fn(_,doc_fn,_)
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
101 doc = doc_fn()
257
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
102 end
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
103 local total_hits = index.advanced_search(query,fn,1)
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
104 total_hits <= 1 or error( "found " .. total_hits .. " documents" )
257
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
105 return doc
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
106 end
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
107
c5c60eca33dd allow Lucene search for 0 rows
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 233
diff changeset
108 function index.count(query)
545
ddcd4296107a clean up lucene search
Franklin Schmidt <fschmidt@gmail.com>
parents: 544
diff changeset
109 return index.advanced_search(query)
232
9ce18106f95a more lucene work
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 230
diff changeset
110 end
9ce18106f95a more lucene work
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents: 230
diff changeset
111
625
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
112 function index.html_highlighter(query,formatter,container_tags)
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
113 local highlighter = index.highlighter(query,formatter)
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
114 return function(html)
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
115 local list = Html.parse(html,container_tags)
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
116 local result = {}
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
117 for _, obj in ipairs(list) do
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
118 if type(obj) == "string" then
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
119 obj = highlighter(obj)
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
120 end
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
121 result[#result+1] = obj
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
122 end
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
123 return Html.to_string(result)
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
124 end
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
125 end
a3c1e11fb6aa rewrite much of Html to be more understandable;
Franklin Schmidt <fschmidt@gmail.com>
parents: 624
diff changeset
126
1367
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
127 function index.backup_to(backup_dir)
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
128 backup_dir.delete()
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
129 backup_dir.mkdir()
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
130 index.snapshot( function(dir_path,file_names)
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
131 local lucene_dir = uri("file:"..dir_path)
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
132 for _, file_name in ipairs(file_names) do
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
133 local lucene_file = lucene_dir.child(file_name)
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
134 local backup_file = backup_dir.child(file_name)
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
135 backup_file.link_to(lucene_file)
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
136 end
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
137 end )
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
138 end
836e00bf7ce2 add Lucene backup_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1365
diff changeset
139
1368
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
140 function index.schedule_backups_to(backup_dir,repeating_delay)
1369
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
141 local lucene_dir = index.dir
1368
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
142 local function backup()
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
143 if backup_dir.last_modified() < lucene_dir.last_modified() then
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
144 run_for_backup(function()
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
145 index.backup_to(backup_dir)
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
146 logger.info "backup"
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
147 end)
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
148 end
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
149 end
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
150 schedule( backup, {
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
151 repeating_delay = repeating_delay or Time.period{minutes=10}
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
152 run_on_close = true
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
153 daemon = true
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
154 } )
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
155 end
5225cd6ed478 add Lucene schedule_backups_to
Franklin Schmidt <fschmidt@gmail.com>
parents: 1367
diff changeset
156
707
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
157 function index.zip(zip_file)
761
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
158 index.snapshot( function(dir_path,file_names)
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
159 zip_file.delete()
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
160 local zip_path = zip_file.canonical().to_string()
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
161 local dir = uri("file:"..dir_path)
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
162 local dir_name = dir.name()
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
163 local options = {dir=dir.parent()}
707
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
164 for _, file_name in ipairs(file_names) do
761
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
165 local cmd = "zip "..zip_path.." "..dir_name.."/"..file_name
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
166 Io.uri("os:"..cmd,options).read_text()
707
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
167 end
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
168 end )
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
169 end
1365
6617763dfd76 fix logging and lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 1353
diff changeset
170 index.zip = Boot.no_security(index.zip)
707
1ed9e55f0be8 replace lucene.backup with lucene.zip implemented in luan, and add lucene.snapshot
Franklin Schmidt <fschmidt@gmail.com>
parents: 693
diff changeset
171
754
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
172 function index.restore(zip_file)
756
9092e52f94eb better synchronization for lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 754
diff changeset
173 java_index.run_in_lock( function()
1369
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
174 local lucene_dir = index.dir
756
9092e52f94eb better synchronization for lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 754
diff changeset
175 local before_restore = lucene_dir.parent().child("before_restore.zip")
9092e52f94eb better synchronization for lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 754
diff changeset
176 index.zip(before_restore)
1345
6f8988830098 unique LuceneIndex per dir
Franklin Schmidt <fschmidt@gmail.com>
parents: 1344
diff changeset
177 java_index.doClose()
756
9092e52f94eb better synchronization for lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 754
diff changeset
178 lucene_dir.delete()
761
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
179 Io.uri("os:unzip "..zip_file.canonical().to_string(),{dir=lucene_dir.parent()}).read_text()
756
9092e52f94eb better synchronization for lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 754
diff changeset
180 java_index.reopen()
9092e52f94eb better synchronization for lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 754
diff changeset
181 end )
754
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
182 end
1365
6617763dfd76 fix logging and lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 1353
diff changeset
183 index.restore = Boot.no_security(index.restore)
754
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
184
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
185 local function multi_error()
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
186 error "multiple lucene instances"
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
187 end
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
188
1133
ba4daf107e07 fix security bug
Franklin Schmidt <fschmidt@gmail.com>
parents: 1129
diff changeset
189 if Rpc.functions.lucene_backup == nil then
754
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
190
749
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
191 function Rpc.functions.lucene_backup(password)
757
Franklin Schmidt <fschmidt@gmail.com>
parents: 756
diff changeset
192 Io.password == password or error "wrong password"
1369
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
193 local zip_file = index.dir.parent().child("backup.zip")
749
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
194 index.zip(zip_file)
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
195 return zip_file
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
196 end
754
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
197
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
198 function Rpc.functions.lucene_restore(password,zip_file)
757
Franklin Schmidt <fschmidt@gmail.com>
parents: 756
diff changeset
199 Io.password == password or error "wrong password"
1369
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
200 local backup_zip = index.dir.parent().child("backup.zip")
761
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
201 backup_zip.write(zip_file)
99356cfde2f0 remove horrible java zip
Franklin Schmidt <fschmidt@gmail.com>
parents: 757
diff changeset
202 index.restore(backup_zip)
754
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
203 end
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
204
749
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
205 else
754
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
206 Rpc.functions.lucene_backup = multi_error
1a101ac9ea46 add lucene restore
Franklin Schmidt <fschmidt@gmail.com>
parents: 753
diff changeset
207 Rpc.functions.lucene_restore = multi_error
749
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
208 end
85f5444fb7d4 add rcp lucene backup
Franklin Schmidt <fschmidt@gmail.com>
parents: 748
diff changeset
209
230
4438cb2e04d0 start lucene
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents:
diff changeset
210 return index
4438cb2e04d0 start lucene
fschmidt@gmail.com <fschmidt@gmail.com@21e917c8-12df-6dd8-5cb6-c86387c605b9>
parents:
diff changeset
211 end
503
92c3d22745b8 make _ENV optional
Franklin Schmidt <fschmidt@gmail.com>
parents: 435
diff changeset
212
1369
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
213 function Lucene.recover(index_dir,restore_dir)
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
214 if not index_dir.exists() and restore_dir.exists() then
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
215 index_dir.mkdir()
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
216 for _, child in ipairs(restore_dir.children()) do
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
217 index_dir.child(child.name()).link_to(child)
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
218 end
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
219 logger.error "recovered"
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
220 end
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
221 end
709f7498a363 change Lucene.index() and add Lucene.recover()
Franklin Schmidt <fschmidt@gmail.com>
parents: 1368
diff changeset
222
1088
bae2d0c2576c change module naming convention
Franklin Schmidt <fschmidt@gmail.com>
parents: 796
diff changeset
223 return Lucene