tsql_parser 0.1.6 → 0.1.8
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/lib/parsing/formatter.rb +8 -8
- data/lib/parsing/formatters/format_factory.rb +1 -0
- data/lib/parsing/formatters/strategy/__formatters.rb +2 -0
- data/lib/parsing/formatters/strategy/cte_formatter.rb +28 -0
- data/lib/parsing/formatters/strategy/from_formatter.rb +46 -0
- data/lib/parsing/formatters/strategy/insert_formatter.rb +1 -0
- data/lib/parsing/formatters/strategy/join_formatter.rb +56 -7
- data/lib/parsing/formatters/strategy/select_formatter.rb +27 -2
- data/lib/parsing/formatters/strategy/set_formatter.rb +2 -1
- data/lib/parsing/formatters/strategy/update_formatter.rb +3 -1
- data/lib/parsing/formatters/strategy/where_formatter.rb +11 -3
- data/lib/parsing/tokenizer.rb +2 -1
- metadata +6 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 9e811afa0fc437610f1466d197aae259f227095dfa7c2d8ebfc6022387984072
|
4
|
+
data.tar.gz: c903812e94a81e2ba40c37f514315e25b86731768af0ce3829ca95879cf98ae9
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: d6b6cd6ccaf82f4267b5028c053f98e6f0960a3a88110d9b623cc3ddc4e770392fb452e7d9fa78126071d89aa742102ecd0e1e49b7030d830c5d39cc6bb49d3a
|
7
|
+
data.tar.gz: a4ed89d5d67b03a7128a102a28edfe2b9b5b46bc115e25e83c14293314170e2cfa3f1c8e5bd182f967aad786c1aabd9d40f7f2c172e097c68a75a33d67d6b417
|
data/lib/parsing/formatter.rb
CHANGED
@@ -25,15 +25,15 @@ module TSqlParser::Parsing
|
|
25
25
|
lines = TokenTransformer.transform(tokens)
|
26
26
|
lines = self.cleanup_whitespace(lines)
|
27
27
|
lines = self.insert_indentation(lines, tab_count, tab)
|
28
|
-
#lines = self.insert_newlines(lines)
|
29
28
|
text = lines.join("\n")
|
30
|
-
text = TextFormatter.new(
|
31
|
-
text = TextFormatter.new(
|
32
|
-
text = TextFormatter.new(
|
33
|
-
text = TextFormatter.new(
|
34
|
-
text = TextFormatter.new(
|
35
|
-
text = TextFormatter.new(
|
36
|
-
|
29
|
+
text = TextFormatter.new(CTE, text, tab).format
|
30
|
+
text = TextFormatter.new(WHERE, text, tab).format if text.include? " WHERE "
|
31
|
+
text = TextFormatter.new(FROM, text, tab).format if text.include? " FROM "
|
32
|
+
text = TextFormatter.new(SELECT, text, tab).format if text.include? " SELECT "
|
33
|
+
text = TextFormatter.new(SET, text, tab).format if text.include? " SET "
|
34
|
+
text = TextFormatter.new(UPDATE, text, tab).format if text.include? " UPDATE "
|
35
|
+
text = TextFormatter.new(JOIN, text, tab).format if text.include? " JOIN "
|
36
|
+
text = TextFormatter.new(INSERT, text, tab).format if text.include? " INSERT "
|
37
37
|
text
|
38
38
|
end
|
39
39
|
|
@@ -20,6 +20,7 @@ module TSqlParser::Parsing
|
|
20
20
|
require_relative "select_formatter"
|
21
21
|
require_relative "update_formatter"
|
22
22
|
require_relative "where_formatter"
|
23
|
+
require_relative "from_formatter"
|
23
24
|
|
24
25
|
CTE = 0
|
25
26
|
INSERT = 1
|
@@ -28,4 +29,5 @@ module TSqlParser::Parsing
|
|
28
29
|
SET = 4
|
29
30
|
UPDATE = 5
|
30
31
|
WHERE = 6
|
32
|
+
FROM = 7
|
31
33
|
end
|
@@ -19,6 +19,34 @@ module TSqlParser::Parsing::Formatters
|
|
19
19
|
|
20
20
|
class CommonTableExpressionFormatter < BaseFormatter
|
21
21
|
def format(text, tab = Defaults.get_default_tab)
|
22
|
+
formatted = []
|
23
|
+
lines = text.split("\n")
|
24
|
+
lines.each_with_index do |line, index|
|
25
|
+
clean_line = line.strip
|
26
|
+
tab_count = self.get_tab_count(line, tab)
|
27
|
+
if clean_line.include? " AS (SELECT "
|
28
|
+
if clean_line.start_with? "WITH" and clean_line.end_with? ")"
|
29
|
+
cte_parts = clean_line.split(" AS (SELECT ")
|
30
|
+
cte_name = cte_parts[0]
|
31
|
+
cte_body = "AS (\n#{tab * (tab_count + 1)}SELECT #{cte_parts[1][..-2]}\n#{tab * tab_count})"
|
32
|
+
formatted << "#{tab * tab_count}#{cte_name} #{cte_body}"
|
33
|
+
elsif clean_line.end_with? ") ,"
|
34
|
+
cte_parts = clean_line.split(" AS (SELECT ")
|
35
|
+
cte_name = cte_parts[0]
|
36
|
+
cte_body = "AS (\n#{tab * (tab_count + 1)}SELECT #{cte_parts[1].sub(") ,", "")}\n#{tab * tab_count}),"
|
37
|
+
formatted << "#{tab * tab_count}#{cte_name} #{cte_body}"
|
38
|
+
elsif clean_line.end_with? ")"
|
39
|
+
cte_parts = clean_line.split(" AS (SELECT ")
|
40
|
+
cte_name = cte_parts[0]
|
41
|
+
cte_body = "AS (\n#{tab * (tab_count + 1)}SELECT #{cte_parts[1][..-2]}\n#{tab * tab_count})"
|
42
|
+
formatted << "#{tab * tab_count}#{cte_name} #{cte_body}"
|
43
|
+
end
|
44
|
+
else
|
45
|
+
formatted << line
|
46
|
+
end
|
47
|
+
end
|
48
|
+
|
49
|
+
formatted.join("\n")
|
22
50
|
end
|
23
51
|
end
|
24
52
|
end
|
@@ -0,0 +1,46 @@
|
|
1
|
+
# __ .__
|
2
|
+
# _/ |_ ___________| | ___________ _______ ______ ___________
|
3
|
+
# \ __\/ ___/ ____/ | ______ \____ \__ \\_ __ \/ ___// __ \_ __ \
|
4
|
+
# | | \___ < <_| | |__ /_____/ | |_> > __ \| | \/\___ \\ ___/| | \/
|
5
|
+
# |__| /____ >__ |____/ | __(____ /__| /____ >\___ >__|
|
6
|
+
# \/ |__| |__| \/ \/ \/
|
7
|
+
#
|
8
|
+
# A very light-weight and opinionated T-SQL parser and formatter.
|
9
|
+
#
|
10
|
+
# github.com/scstauf
|
11
|
+
#
|
12
|
+
# path:
|
13
|
+
# parsing/formatters/from_formatter.rb
|
14
|
+
# object:
|
15
|
+
# TSqlParser::Parsing::Formatters::FromFormatter
|
16
|
+
|
17
|
+
module TSqlParser::Parsing::Formatters
|
18
|
+
require_relative "base_formatter"
|
19
|
+
|
20
|
+
class FromFormatter < BaseFormatter
|
21
|
+
def format(text, tab = Defaults.get_default_tab)
|
22
|
+
formatted = []
|
23
|
+
lines = text.split("\n")
|
24
|
+
lines.each_with_index do |line, index|
|
25
|
+
if line.strip.start_with? "--" or line.strip.start_with? "/*" or line.strip.end_with? "'"
|
26
|
+
formatted << line
|
27
|
+
next
|
28
|
+
end
|
29
|
+
|
30
|
+
if line.include? " FROM " and not line.include? " EXISTS ("
|
31
|
+
tab_count = self.get_tab_count(line, tab)
|
32
|
+
if line.count(" FROM ") == 1
|
33
|
+
from_parts = line.split(" FROM ")
|
34
|
+
formatted << "#{from_parts[0]}\n#{tab * tab_count}FROM #{from_parts[1]}"
|
35
|
+
else
|
36
|
+
formatted << line[..line.index(" FROM ") - 1]
|
37
|
+
formatted << "#{tab * tab_count}#{line[line.index(" FROM ") + 1..]}"
|
38
|
+
end
|
39
|
+
else
|
40
|
+
formatted << line
|
41
|
+
end
|
42
|
+
end
|
43
|
+
formatted.join("\n")
|
44
|
+
end
|
45
|
+
end
|
46
|
+
end
|
@@ -28,13 +28,17 @@ module TSqlParser::Parsing::Formatters
|
|
28
28
|
|
29
29
|
lines.each do |line|
|
30
30
|
first = line.strip.split(" ").first
|
31
|
-
if
|
32
|
-
|
33
|
-
|
34
|
-
|
35
|
-
|
36
|
-
|
37
|
-
|
31
|
+
next if first.nil?
|
32
|
+
|
33
|
+
if first.start_with? "--" or first.start_with? "/*" or line.strip.end_with? "'"
|
34
|
+
new_text << line
|
35
|
+
next
|
36
|
+
end
|
37
|
+
|
38
|
+
tab_count = self.get_tab_count(line, tab)
|
39
|
+
|
40
|
+
if line.include? " JOIN "
|
41
|
+
new_text << self.format_joins(line, tab, tab_count)
|
38
42
|
else
|
39
43
|
new_text << line
|
40
44
|
end
|
@@ -42,5 +46,50 @@ module TSqlParser::Parsing::Formatters
|
|
42
46
|
|
43
47
|
new_text.join("\n")
|
44
48
|
end
|
49
|
+
|
50
|
+
private
|
51
|
+
|
52
|
+
def format_joins(line, tab, tab_count)
|
53
|
+
formatted = []
|
54
|
+
builder = []
|
55
|
+
tokens = line.strip.split(" ")
|
56
|
+
skip_count = 0
|
57
|
+
tokens.each_with_index do |t, i|
|
58
|
+
if skip_count > 0
|
59
|
+
skip_count -= 1
|
60
|
+
next
|
61
|
+
end
|
62
|
+
|
63
|
+
last_one = tokens[i - 1] if i - 1 > 0
|
64
|
+
future_one = tokens[i + 1] if i + 1 < tokens.size
|
65
|
+
future_two = tokens[i + 2] if i + 2 < tokens.size
|
66
|
+
future_three = tokens[i + 3] if i + 3 < tokens.size
|
67
|
+
|
68
|
+
next_two = "#{future_one} #{future_two}"
|
69
|
+
next_three = "#{future_one} #{future_two} #{future_three}"
|
70
|
+
|
71
|
+
if ["INNER JOIN", "LEFT JOIN", "RIGHT JOIN", "FULL JOIN", "CROSS JOIN"].include? next_two
|
72
|
+
builder << t
|
73
|
+
formatted << "#{tab * tab_count}#{builder.join(" ")}" unless builder.empty?
|
74
|
+
builder = [next_two]
|
75
|
+
skip_count = 2
|
76
|
+
elsif ["LEFT OUTER JOIN", "RIGHT OUTER JOIN", "FULL OUTER JOIN"].include? next_three
|
77
|
+
builder << t
|
78
|
+
formatted << "#{tab * tab_count}#{builder.join(" ")}" unless builder.empty?
|
79
|
+
builder = [next_three]
|
80
|
+
skip_count = 3
|
81
|
+
elsif t == "JOIN" and (not last_one.nil? and not %w[INNER LEFT RIGHT FULL CROSS OUTER].include? last_one)
|
82
|
+
formatted << "#{tab * tab_count}#{builder.join(" ")}" unless builder.empty?
|
83
|
+
builder = [t]
|
84
|
+
else
|
85
|
+
builder << t
|
86
|
+
end
|
87
|
+
end
|
88
|
+
|
89
|
+
formatted << "#{tab * tab_count}#{builder.join(" ")}" unless builder.empty?
|
90
|
+
builder = []
|
91
|
+
|
92
|
+
formatted.join("\n")
|
93
|
+
end
|
45
94
|
end
|
46
95
|
end
|
@@ -23,6 +23,8 @@ module TSqlParser::Parsing::Formatters
|
|
23
23
|
lines = text.split("\n")
|
24
24
|
lines.each do |line|
|
25
25
|
first = line.strip.split(" ").first
|
26
|
+
next if first.nil?
|
27
|
+
|
26
28
|
if first != "SELECT"
|
27
29
|
formatted << line
|
28
30
|
next
|
@@ -45,8 +47,31 @@ module TSqlParser::Parsing::Formatters
|
|
45
47
|
def format_select(s, tab_count = Defaults.get_default_tab_count, tab = Defaults.get_default_tab)
|
46
48
|
return s if s.nil?
|
47
49
|
|
48
|
-
|
49
|
-
|
50
|
+
new_tokens = []
|
51
|
+
parenthesis = 0
|
52
|
+
builder = ""
|
53
|
+
skip_count = 0
|
54
|
+
s.split("").each do |c|
|
55
|
+
parenthesis += 1 if c == "("
|
56
|
+
parenthesis -= 1 if c == ")"
|
57
|
+
|
58
|
+
if skip_count > 0
|
59
|
+
skip_count -= 1
|
60
|
+
next
|
61
|
+
end
|
62
|
+
|
63
|
+
if c == "," and parenthesis == 0
|
64
|
+
new_tokens << builder unless builder.empty?
|
65
|
+
builder = ""
|
66
|
+
skip_count = 1
|
67
|
+
else
|
68
|
+
builder += c
|
69
|
+
end
|
70
|
+
end
|
71
|
+
|
72
|
+
new_tokens << builder unless builder.empty?
|
73
|
+
|
74
|
+
"\n#{new_tokens.map { |t| "#{tab * (tab_count + 1)}#{t}" }.join(",\n")}"
|
50
75
|
end
|
51
76
|
end
|
52
77
|
end
|
@@ -28,6 +28,7 @@ module TSqlParser::Parsing::Formatters
|
|
28
28
|
lines.each do |line|
|
29
29
|
tokens = line.strip.split(" ")
|
30
30
|
first = tokens.first
|
31
|
+
next if first.nil?
|
31
32
|
next_token = tokens[1] if tokens.size > 1
|
32
33
|
|
33
34
|
if %w[FROM WHERE].include? first and wait
|
@@ -59,7 +60,7 @@ module TSqlParser::Parsing::Formatters
|
|
59
60
|
parts = line.strip.split(" SET ")
|
60
61
|
tab_count = self.get_tab_count(line, tab)
|
61
62
|
formatted << "#{tab * tab_count}#{parts[0]}\n"
|
62
|
-
parts[1..].each { |p| formatted << "#{tab * tab_count}SET #{p}" }
|
63
|
+
parts[1..].each { |p| formatted << "#{tab * tab_count}SET #{self.format_set(p, tab_count, tab)}" }
|
63
64
|
elsif wait
|
64
65
|
set_lines << line
|
65
66
|
else
|
@@ -23,6 +23,8 @@ module TSqlParser::Parsing::Formatters
|
|
23
23
|
lines = text.split("\n")
|
24
24
|
lines.each do |line|
|
25
25
|
first = line.strip.split(" ").first
|
26
|
+
next if first.nil?
|
27
|
+
|
26
28
|
if first != "UPDATE"
|
27
29
|
formatted << line
|
28
30
|
next
|
@@ -35,7 +37,7 @@ module TSqlParser::Parsing::Formatters
|
|
35
37
|
formatted << line
|
36
38
|
next
|
37
39
|
end
|
38
|
-
formatted << line.sub(update, new_update)
|
40
|
+
formatted << "#{tab * tab_count}#{line.strip.sub(update, new_update)}"
|
39
41
|
end
|
40
42
|
formatted.join("\n")
|
41
43
|
end
|
@@ -22,19 +22,27 @@ module TSqlParser::Parsing::Formatters
|
|
22
22
|
formatted = []
|
23
23
|
text.split("\n").each do |line|
|
24
24
|
first = line.strip.split(" ").first
|
25
|
-
if first
|
25
|
+
next if first.nil?
|
26
|
+
if not line.include? "WHERE"
|
26
27
|
formatted << line
|
27
28
|
next
|
28
29
|
end
|
29
30
|
|
31
|
+
if line.strip.start_with? "--" or line.strip.start_with? "/*" or line.strip.end_with? "'"
|
32
|
+
formatted << line
|
33
|
+
next
|
34
|
+
end
|
35
|
+
|
36
|
+
where_clause = line[line.index(" WHERE ")..].strip
|
37
|
+
|
30
38
|
tab_count = self.get_tab_count(line, tab)
|
31
|
-
predicate =
|
39
|
+
predicate = where_clause.strip["WHERE".size + 1..]
|
32
40
|
new_predicate = self.format_predicate(predicate, tab_count, tab)
|
33
41
|
if new_predicate.nil?
|
34
42
|
formatted << line
|
35
43
|
next
|
36
44
|
end
|
37
|
-
formatted << line.sub(predicate, new_predicate)
|
45
|
+
formatted << line.sub("WHERE", "").sub(predicate, "\n#{tab * tab_count}WHERE#{new_predicate}")
|
38
46
|
end
|
39
47
|
|
40
48
|
formatted.join("\n")
|
data/lib/parsing/tokenizer.rb
CHANGED
@@ -25,6 +25,7 @@ module TSqlParser::Parsing
|
|
25
25
|
|
26
26
|
def basic_tokenize(tsql_string)
|
27
27
|
self.reset
|
28
|
+
tsql_string = tsql_string.gsub("\t", Defaults.get_default_tab)
|
28
29
|
tsql_chars = tsql_string.split("")
|
29
30
|
|
30
31
|
tsql_chars.each_with_index do |c, i|
|
@@ -64,7 +65,7 @@ module TSqlParser::Parsing
|
|
64
65
|
def handle_multicomment_start
|
65
66
|
if Parser.is_multiline_comment_start?(@c, @next_c)
|
66
67
|
@multiline_comment = true
|
67
|
-
self.flush_builder(c)
|
68
|
+
self.flush_builder(@c)
|
68
69
|
return true
|
69
70
|
end
|
70
71
|
end
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: tsql_parser
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.8
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Scott Stauffer
|
@@ -10,7 +10,10 @@ bindir: bin
|
|
10
10
|
cert_chain: []
|
11
11
|
date: 2023-04-03 00:00:00.000000000 Z
|
12
12
|
dependencies: []
|
13
|
-
description:
|
13
|
+
description: |-
|
14
|
+
A very light-weight and opinionated T-SQL parser and formatter.
|
15
|
+
Github has the most recent docs.
|
16
|
+
Contributions welcome!
|
14
17
|
email: scott@fuseraft.com
|
15
18
|
executables: []
|
16
19
|
extensions: []
|
@@ -23,6 +26,7 @@ files:
|
|
23
26
|
- lib/parsing/formatters/strategy/__formatters.rb
|
24
27
|
- lib/parsing/formatters/strategy/base_formatter.rb
|
25
28
|
- lib/parsing/formatters/strategy/cte_formatter.rb
|
29
|
+
- lib/parsing/formatters/strategy/from_formatter.rb
|
26
30
|
- lib/parsing/formatters/strategy/insert_formatter.rb
|
27
31
|
- lib/parsing/formatters/strategy/join_formatter.rb
|
28
32
|
- lib/parsing/formatters/strategy/select_formatter.rb
|