ruby-changes:72010
From: Jeremy <ko1@a...>
Date: Mon, 30 May 2022 13:10:57 +0900 (JST)
Subject: [ruby-changes:72010] 609d73e892 (master): [ruby/stringio] Fix handling of chomp with paragraph separator
https://git.ruby-lang.org/ruby.git/commit/?id=609d73e892 From 609d73e8925f807786686caf635178bb1de74256 Mon Sep 17 00:00:00 2001 From: Jeremy Evans <code@j...> Date: Thu, 26 May 2022 15:01:44 -0700 Subject: [ruby/stringio] Fix handling of chomp with paragraph separator Try to mirror IO behavior, where chomp takes out the entire paragraph separators between entries, but does not chomp a single line separator at the end of the string. Partially Fixes [Bug #18768] https://github.com/ruby/stringio/commit/a83ddbb7f0 --- ext/stringio/stringio.c | 27 +++++++++++++++------------ test/stringio/test_stringio.rb | 24 ++++++++++++------------ 2 files changed, 27 insertions(+), 24 deletions(-) diff --git a/ext/stringio/stringio.c b/ext/stringio/stringio.c index 3f66fb17d2..13c8af9216 100644 --- a/ext/stringio/stringio.c +++ b/ext/stringio/stringio.c @@ -1204,6 +1204,7 @@ strio_getline(struct getline_arg *arg, struct StringIO *ptr) https://github.com/ruby/ruby/blob/trunk/ext/stringio/stringio.c#L1204 str = strio_substr(ptr, ptr->pos, e - s - w, enc); } else if ((n = RSTRING_LEN(str)) == 0) { + const char *paragraph_end = NULL; p = s; while (p[(p + 1 < e) && (*p == '\r') && 0] == '\n') { p += *p == '\r'; @@ -1213,19 +1214,21 @@ strio_getline(struct getline_arg *arg, struct StringIO *ptr) https://github.com/ruby/ruby/blob/trunk/ext/stringio/stringio.c#L1214 } s = p; while ((p = memchr(p, '\n', e - p)) && (p != e)) { - if (*++p == '\n') { - e = p + 1; - w = (arg->chomp ? 1 : 0); - break; - } - else if (*p == '\r' && p < e && p[1] == '\n') { - e = p + 2; - w = (arg->chomp ? 2 : 0); - break; - } + p++; + if (!((p < e && *p == '\n') || + (p + 1 < e && *p == '\r' && *(p+1) == '\n'))) { + continue; + } + paragraph_end = p - ((*(p-2) == '\r') ? 2 : 1); + while ((p < e && *p == '\n') || + (p + 1 < e && *p == '\r' && *(p+1) == '\n')) { + p += (*p == '\r') ? 2 : 1; + } + e = p; + break; } - if (!w && arg->chomp) { - w = chomp_newline_width(s, e); + if (arg->chomp && paragraph_end) { + w = e - paragraph_end; } str = strio_substr(ptr, s - RSTRING_PTR(ptr->string), e - s - w, enc); } diff --git a/test/stringio/test_stringio.rb b/test/stringio/test_stringio.rb index f0af708c10..5393162573 100644 --- a/test/stringio/test_stringio.rb +++ b/test/stringio/test_stringio.rb @@ -93,10 +93,10 @@ class TestStringIO < Test::Unit::TestCase https://github.com/ruby/ruby/blob/trunk/test/stringio/test_stringio.rb#L93 assert_equal("a", StringIO.new("a\nb").gets(chomp: true)) assert_equal("abc", StringIO.new("abc\n\ndef\n").gets(chomp: true)) assert_equal("abc\n\ndef\n", StringIO.new("abc\n\ndef\n").gets(nil, chomp: true)) - assert_equal("abc\n", StringIO.new("abc\n\ndef\n").gets("", chomp: true)) + assert_equal("abc", StringIO.new("abc\n\ndef\n").gets("", chomp: true)) stringio = StringIO.new("abc\n\ndef\n") - assert_equal("abc\n", stringio.gets("", chomp: true)) - assert_equal("def", stringio.gets("", chomp: true)) + assert_equal("abc", stringio.gets("", chomp: true)) + assert_equal("def\n", stringio.gets("", chomp: true)) assert_string("", Encoding::UTF_8, StringIO.new("\n").gets(chomp: true)) end @@ -110,10 +110,10 @@ class TestStringIO < Test::Unit::TestCase https://github.com/ruby/ruby/blob/trunk/test/stringio/test_stringio.rb#L110 assert_equal("a", StringIO.new("a\r\nb").gets(chomp: true)) assert_equal("abc", StringIO.new("abc\r\n\r\ndef\r\n").gets(chomp: true)) assert_equal("abc\r\n\r\ndef\r\n", StringIO.new("abc\r\n\r\ndef\r\n").gets(nil, chomp: true)) - assert_equal("abc\r\n", StringIO.new("abc\r\n\r\ndef\r\n").gets("", chomp: true)) + assert_equal("abc", StringIO.new("abc\r\n\r\ndef\r\n").gets("", chomp: true)) stringio = StringIO.new("abc\r\n\r\ndef\r\n") - assert_equal("abc\r\n", stringio.gets("", chomp: true)) - assert_equal("def", stringio.gets("", chomp: true)) + assert_equal("abc", stringio.gets("", chomp: true)) + assert_equal("def\r\n", stringio.gets("", chomp: true)) end def test_readlines @@ -596,15 +596,15 @@ class TestStringIO < Test::Unit::TestCase https://github.com/ruby/ruby/blob/trunk/test/stringio/test_stringio.rb#L596 assert_equal(["foo\n", "bar\n", "baz\n"], f.each.to_a) f.rewind assert_equal(["foo", "bar", "baz"], f.each(chomp: true).to_a) - f = StringIO.new("foo\nbar\n\nbaz\n") - assert_equal(["foo\nbar\n\n", "baz\n"], f.each("").to_a) + f = StringIO.new("foo\nbar\n\n\nbaz\n") + assert_equal(["foo\nbar\n\n\n", "baz\n"], f.each("").to_a) f.rewind - assert_equal(["foo\nbar\n", "baz"], f.each("", chomp: true).to_a) + assert_equal(["foo\nbar", "baz\n"], f.each("", chomp: true).to_a) - f = StringIO.new("foo\r\nbar\r\n\r\nbaz\r\n") - assert_equal(["foo\r\nbar\r\n\r\n", "baz\r\n"], f.each("").to_a) + f = StringIO.new("foo\r\nbar\r\n\r\n\r\nbaz\r\n") + assert_equal(["foo\r\nbar\r\n\r\n\r\n", "baz\r\n"], f.each("").to_a) f.rewind - assert_equal(["foo\r\nbar\r\n", "baz"], f.each("", chomp: true).to_a) + assert_equal(["foo\r\nbar", "baz\r\n"], f.each("", chomp: true).to_a) f = StringIO.new("abc\n\ndef\n") assert_equal(["ab", "c\n", "\nd", "ef", "\n"], f.each(nil, 2, chomp: true).to_a) -- cgit v1.2.1 -- ML: ruby-changes@q... Info: http://www.atdot.net/~ko1/quickml/