Skip to content

Raise IndexError on unmatched subpattern for MatchData#begin and #end #12810

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
straight-shoota merged 2 commits into from
Dec 2, 2022
Merged

Raise IndexError on unmatched subpattern for MatchData#begin and #end #12810

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
89f3761
Remove duplicate regex evaluation in `MatchData` spec
straight-shoota Nov 30, 2022
58221bc
Raise `IndexError` on unmatched subpattern for `MatchData#begin` and …
straight-shoota Nov 30, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
102 changes: 90 additions & 12 deletions spec/std/regex/match_data_spec.cr
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,12 +65,33 @@ describe "Regex::MatchData" do
end

it "with capture" do
matchdata(/f(o)o/, "foo").begin.should eq 0
matchdata(/f(o)o/, "foo").begin(1).should eq 1
matchdata(/f(o)o/, "foo").begin(-1).should eq 1
matchdata(/f(o)o/, ".foo.").begin.should eq 1
matchdata(/f(o)o/, ".foo.").begin(1).should eq 2
matchdata(/f(o)o/, ".foo.").begin(-1).should eq 2
md = matchdata(/f(o)o/, "foo")
md.begin.should eq 0
md.begin(1).should eq 1
md.begin(-1).should eq 1

md = matchdata(/f(o)o/, ".foo.")
md.begin.should eq 1
md.begin(1).should eq 2
md.begin(-1).should eq 2
end

it "with unmatched capture" do
md = matchdata(/f(x)?o/, "foo")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.begin(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.begin(-1)
end

md = matchdata(/f(x)?o/, ".foo.")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.begin(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.begin(-1)
end
end

it "char index" do
Expand All @@ -82,6 +103,24 @@ describe "Regex::MatchData" do
it "char index" do
matchdata(/foo/, "öfoo").byte_begin.should eq 2
end

it "with unmatched capture" do
md = matchdata(/f(x)?o/, "foo")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_begin(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_begin(-1)
end

md = matchdata(/f(x)?o/, ".foo.")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_begin(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_begin(-1)
end
end
end

describe "#end" do
Expand All @@ -99,12 +138,33 @@ describe "Regex::MatchData" do
end

it "with capture" do
matchdata(/f(o)o/, "foo").end.should eq 3
matchdata(/f(o)o/, "foo").end(1).should eq 2
matchdata(/f(o)o/, "foo").end(-1).should eq 2
matchdata(/f(o)o/, ".foo.").end.should eq 4
matchdata(/f(o)o/, ".foo.").end(1).should eq 3
matchdata(/f(o)o/, ".foo.").end(-1).should eq 3
md = matchdata(/f(o)o/, "foo")
md.end.should eq 3
md.end(1).should eq 2
md.end(-1).should eq 2

md = matchdata(/f(o)o/, ".foo.")
md.end.should eq 4
md.end(1).should eq 3
md.end(-1).should eq 3
end

it "with unmatched capture" do
md = matchdata(/f(x)?o/, "foo")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.end(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.end(-1)
end

md = matchdata(/f(x)?o/, ".foo.")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.end(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.end(-1)
end
end

it "char index" do
Expand All @@ -116,6 +176,24 @@ describe "Regex::MatchData" do
it "char index" do
matchdata(/foo/, "öfoo").byte_end.should eq 5
end

it "with unmatched capture" do
md = matchdata(/f(x)?o/, "foo")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_end(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_end(-1)
end

md = matchdata(/f(x)?o/, ".foo.")
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_end(1)
end
expect_raises(IndexError, "Capture group 1 was not matched") do
md.byte_end(-1)
end
end
end

describe "#[]" do
Expand Down
28 changes: 26 additions & 2 deletions src/regex/match_data.cr
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,10 +59,15 @@ class Regex
# When *n* is `0` or not given, uses the match of the entire `Regex`.
# Otherwise, uses the match of the *n*th capture group.
#
# Raises `IndexError` if the index is out of range or the respective
# subpattern is unused.
#
# ```
# "Crystal".match(/r/).not_nil!.begin(0) # => 1
# "Crystal".match(/r(ys)/).not_nil!.begin(1) # => 2
# "クリスタル".match(/リ(ス)/).not_nil!.begin(0) # => 1
# "Crystal".match(/r/).not_nil!.begin(1) # IndexError: Invalid capture group index: 1
# "Crystal".match(/r(x)?/).not_nil!.begin(1) # IndexError: Capture group 1 was not matched
# ```
def begin(n = 0) : Int32
@string.byte_index_to_char_index(byte_begin(n)).not_nil!
Expand All @@ -73,10 +78,15 @@ class Regex
# When *n* is `0` or not given, uses the match of the entire `Regex`.
# Otherwise, uses the match of the *n*th capture group.
#
# Raises `IndexError` if the index is out of range or the respective
# subpattern is unused.
#
# ```
# "Crystal".match(/r/).not_nil!.end(0) # => 2
# "Crystal".match(/r(ys)/).not_nil!.end(1) # => 4
# "クリスタル".match(/リ(ス)/).not_nil!.end(0) # => 3
# "Crystal".match(/r/).not_nil!.end(1) # IndexError: Invalid capture group index: 1
# "Crystal".match(/r(x)?/).not_nil!.end(1) # IndexError: Capture group 1 was not matched
# ```
def end(n = 0) : Int32
@string.byte_index_to_char_index(byte_end(n)).not_nil!
Expand All @@ -87,31 +97,45 @@ class Regex
# When *n* is `0` or not given, uses the match of the entire `Regex`.
# Otherwise, uses the match of the *n*th capture group.
#
# Raises `IndexError` if the index is out of range or the respective
# subpattern is unused.
#
# ```
# "Crystal".match(/r/).not_nil!.byte_begin(0) # => 1
# "Crystal".match(/r(ys)/).not_nil!.byte_begin(1) # => 2
# "クリスタル".match(/リ(ス)/).not_nil!.byte_begin(0) # => 3
# "Crystal".match(/r/).not_nil!.byte_begin(1) # IndexError: Invalid capture group index: 1
# "Crystal".match(/r(x)?/).not_nil!.byte_begin(1) # IndexError: Capture group 1 was not matched
# ```
def byte_begin(n = 0) : Int32
check_index_out_of_bounds n
n += size if n < 0
@ovector[n * 2]
value = @ovector[n * 2]
raise_capture_group_was_not_matched(n) if value < 0
value
end

# Returns the position of the next byte after the match.
#
# When *n* is `0` or not given, uses the match of the entire `Regex`.
# Otherwise, uses the match of the *n*th capture group.
#
# Raises `IndexError` if the index is out of range or the respective
# subpattern is unused.
#
# ```
# "Crystal".match(/r/).not_nil!.byte_end(0) # => 2
# "Crystal".match(/r(ys)/).not_nil!.byte_end(1) # => 4
# "クリスタル".match(/リ(ス)/).not_nil!.byte_end(0) # => 9
# "Crystal".match(/r/).not_nil!.byte_end(1) # IndexError: Invalid capture group index: 1
# "Crystal".match(/r(x)?/).not_nil!.byte_end(1) # IndexError: Capture group 1 was not matched
# ```
def byte_end(n = 0) : Int32
check_index_out_of_bounds n
n += size if n < 0
@ovector[n * 2 + 1]
value = @ovector[n * 2 + 1]
raise_capture_group_was_not_matched(n) if value < 0
value
end

# Returns the match of the *n*th capture group, or `nil` if there isn't
Expand Down