tools: improve changelog helper; finish 0.4.1 changelogs

CHANGELOG.md

@@ -2,16 +2,18 @@
 *3 September 2023*
 
 #### Improvements in the language
-- Do not allow uninitialized function pointers
-- Fix compiling V programs with latest clang 16 on windows (clang 16 is stricter than clang 14) (#19095)
-ast, checker, cgen: implement enum static `from_string(name string)` (#19156)
 - Pure `array.sorted()` and `array.sorted_with_compare()` methods, that do not modify their receivers (#19251)
 - UB overflow has been removed
+- Implement `Enum.from_string(name string)` for converting strings to enum values (#19156)
 - Disallow casting string to enum, suggest using Enum.from_string() instead (#19260)
+- Use autocasting in complex conditions (#18839)
+- Allow alias as fixed array on return (#18817)
+- Do not allow uninitialized function pointers
+- Fix compiling V programs with latest clang 16 on windows (clang 16 is stricter than clang 14) (#19095)
 - Fix anonymous struct with default expr (#19257)
-- log: improve the most common use case, it's no longer necessary to create a `Log` instance (#19242)
 - Allow using consts as enum values (#19193)
-- `@[attr]` syntax to replace `[attr]` (`[]` is used for too many things). Most likely to be replaced with `@attr()` in the futre.
+- `@[attr]` syntax to replace `[attr]` (`[]` is used for too many things). Most likely to be replaced with `@attr()` in the future.
+- Allow `none` for not first values on map initialization (#18821)
 - Make all .trace() methods generic on the type of the passed expression
 
 #### Breaking changes
@@ -47,26 +49,35 @@ ast, checker, cgen: implement enum static `from_string(name string)` (#19156)
 - Disallow an empty `chan` type (#19167)
 
 #### Parser improvements
-- parser: change warn to error, for const names with upper letter (fix #18838) (#18840)
+- Change warn to error, for const names with upper letter (fix #18838) (#18840)
-- parser: disallow declaring static functions as method receivers (#19007)
+- Disallow declaring static functions as method receivers (#19007)
-- parser: disallow having builtin type as type names for `enum`, `sum type` and `alias` (#19043)
+- Disallow having builtin type as type names for `enum`, `sum type` and `alias` (#19043)
-- parser: support `const x := 123`, to make extracting locals as constants less annoying while prototyping
+- Support `const x := 123`, to make extracting locals as constants less annoying while prototyping
-- parser: fix struct field fn type with default value (fix #19099) (#19106)
+- Fix struct field fn type with default value (fix #19099) (#19106)
-- parser, cgen: fix `for i++; i<10; i++ {` (fix #18445) (#19035)
+- Fix `for i++; i<10; i++ {` (fix #18445) (#19035)
-- parser, checker, cgen: fix fn return alias of fixed array (#19116)
+- Fix fn return alias of fixed array (#19116)
-- Ast, parser, cgen: fix generic struct init (Stack[&Person]{}) (fix #19119) (#19122)
+- Fix generic struct init (Stack[&Person]{}) (fix #19119) (#19122)
-- v.token: add inline next_to() and cleanup related calls (#19226)
 
 #### Compiler internals
+- pref: make -autofree work without -gc none
+- builder,pref: allow thirdparty objects compilation with CPP compiler (#19124)
 - scanner: fix string interpolation with nested string interpolation in inner quotes p. 3 (#19121)
 - scanner: error early on an unsupported escape sequence in a string, like `\_` (fix #19131) (#19134)
+- v.token: add inline next_to() and cleanup related calls (#19226)
 
 #### Standard library
+- eventbus: add generic support for event name (#18805)
+- readline: add support for ctlr+w and ctrl+u shortcuts (#18921)
+- strconv.atoi: fix string.int() returning numbers for non number characters (fix #18875) (#18925)
+- builtin: reduce the number of array allocations for consts in all V programs (#18889)
+- builtin: move array.data to the top of the struct
+- os.notify: implement the kqueue backend for notify.FdNotifier (#19057)
 - vlib: add a new module `builtin.wchar`, to ease dealing with C APIs that accept `wchar_t*` (#18794)
 - arrays: add more util functions and tests for them - find_first, find_last, join_to_string (#18784)
 - vlib: use sync.new_mutex() consistently for initialising all vlib structures containing mutex fields
 - crypto.pem: add a static method `Block.new`, to replace `new` (#18846)
 - crypto.pem: add decode_only and general improvements to decoding (#18908)
+- log: improve the most common use case, it's no longer necessary to create a `Log` instance (#19242)
 - crypto.sha512: make the new384/0, new512_256/0, new512_224/0 functions public
 - json: fix option alias support (#18801)
 - time: fix `parse_format` with `YY` (#18887)
@@ -90,6 +101,14 @@ ast, checker, cgen: implement enum static `from_string(name string)` (#19156)
 - toml: Add generic automatic decoding and encoding of simple structs, when they don't implement custom methods (#17970)
 
 #### Web
+- http: Request.host
+- net.ftp: fix dir() for file names, which contain spaces (fix #18800) (#18804)
+- net.http: make listener public, and add addr in Server struct (#18871)
+- net.http.chunked: return `!string` on decode (#18928)
+- net.conv: rename functions to match other langs, making them easier t… (#18937)
+- wasm: remove dependency on thirdparty/binaryen, webassembly backend rewrite (#18120)
+- wasm: add a -wasm-stack-top flag to compiler (#19157)
+- net.mbedtls: add SSLListener to allow creating SSL servers (#19022)
 - picoev, picohttparser: reimplement in V (#18506)
 - vweb: fix parsing of form fields, send with multipart/form-data (by JS fetch)
 - vweb: make vweb route paths case sensitive (#18973)
@@ -98,8 +117,8 @@ ast, checker, cgen: implement enum static `from_string(name string)` (#19156)
 - vweb: add a user_agent utility method to the vweb context (#19204)
 - vweb: avoid the controllers having to be defined in specific order (#19182)
 
-
 #### ORM
+- orm: fix inserting sequential values (id=0), in tables with an i64 primary field (#18791)
 - Add OR in where on update and delete (#19172)
 
 #### Database drivers
@@ -109,6 +128,7 @@ ast, checker, cgen: implement enum static `from_string(name string)` (#19156)
 - db.sqlite: make functions return results, breaking change (#19093)
 
 #### Native backend
+- native: move functions out of amd64.v (#18857)
 
 #### C backend
 - Fix selector code to use interface method table on closure when needed (#18736)
@@ -134,9 +154,9 @@ ast, checker, cgen: implement enum static `from_string(name string)` (#19156)
 - Fix comptime assign with generic result return type (#19192)
 - Fix match with comptime if expr in branch (#19189)
 
-#### Comptime
-
 #### Tools
+- ci: add v-analyzer builds (#18835)
+- ci: cleanup more the contents of the generated v_linux.zip, v_macos.zip, and v_windows.zip, use -skip-unused
 - tools: fix vcomplete for zsh (#18950)
 - tools: support a toc for projects, with single exposing module, in `v doc` (#19001)
 - Add support for `v should-compile-all -c examples/`, which will delete all the produced executables at the end
@@ -150,26 +170,6 @@ ast, checker, cgen: implement enum static `from_string(name string)` (#19156)
 - v.builder: show the number of files, types, modules, when a program is compiled with -stats
 - Improve the output of parser_speed.v and scanner_speed.v
 
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
 
 ## V 0.4
 *1 July 2023*

cmd/tools/changelog_helper.v

@@ -18,12 +18,9 @@ enum Category {
 	comptime
 	tools
 	compiler_internals
+	examples
 }
 
-//__global (
-// lines []Line
-//)
-
 struct Line {
 	category Category
 	text     string
@@ -31,10 +28,11 @@ struct Line {
 
 const log_txt = 'log.txt'
 
-__global (
+struct App {
-	counter     = 0
+	total_lines int
-	total_lines = 0
+mut:
-)
+	counter int
+}
 
 fn main() {
 	if !os.exists(log_txt) {
@@ -42,14 +40,16 @@ fn main() {
 	}
 	lines := os.read_lines(log_txt)!
 	changelog_txt := os.read_file('CHANGELOG.md')!.to_lower()
+	mut app := &App{
+		total_lines: lines.len
+	}
 	// mut counter := 0 // to display how many commits are left
 	for line in lines {
 		s := line.trim_space()
 		if s == '' {
-			counter++
+			app.counter++
 		}
 	}
-	total_lines = lines.len
 	// println('${counter} / ${lines.len}')
 	for line in lines {
 		s := line.to_lower()
@@ -60,12 +60,21 @@ fn main() {
 			continue
 		}
 
-		process_line(line)!
+		app.process_line(line)!
 	}
 	println('done.')
 }
 
-fn process_line(text string) ! {
+fn (mut app App) process_line(text string) ! {
+	if text == '' {
+		return
+	}
+	semicolon_pos := text.index(': ') or {
+		println('no `:` in commit, skipping: "${text}"')
+		return
+	}
+	prefix := text[..semicolon_pos]
+	// Get category based on keywords in the commit message/prefix
 	mut category := Category.improvements
 	if text.contains('checker:') {
 		category = .checker
@@ -75,7 +84,9 @@ fn process_line(text string) ! {
 		category = .db
 	} else if is_stdlib(text) {
 		category = .stdlib
-	} else if text.contains('vweb:') {
+	} else if is_orm(text) {
+		category = .orm
+	} else if is_web(text) {
 		category = .web
 	} else if is_tools(text) {
 		category = .tools
@@ -85,7 +96,13 @@ fn process_line(text string) ! {
 		category = .compiler_internals
 	} else if is_improvements(text) {
 		category = .improvements
+	} else if is_native(text) {
+		category = .native
+	} else if is_examples(text) {
+		// TODO maybe always skip these as well?
+		category = .examples
 	} else if text.contains('docs:') || text.contains('doc:') {
+		// Always skip docs
 		delete_processed_line_from_log(text)!
 		return
 	}
@@ -95,12 +112,7 @@ fn process_line(text string) ! {
 	}
 
 	// Trim everything to the left of `:` for some commits (e.g. `checker: `)
-	semicolon_pos := text.index(': ') or {
-		println('no : in commit, skipping')
-		return
-	}
 	mut s := text
-	prefix := text[..semicolon_pos]
 	// println("PREFIX='${prefix}'")
 	// if true {
 	// exit(0)
@@ -111,7 +123,7 @@ fn process_line(text string) ! {
 
 	// Get input from the user
 	print('\033[H\033[J')
-	println('${counter} / ${total_lines}')
+	println('${app.counter} / ${app.total_lines}')
 	// println('\n')
 	println(text)
 	input := os.input('${category} ?')
@@ -149,7 +161,7 @@ fn process_line(text string) ! {
 		}
 		else {}
 	}
-	counter++
+	app.counter++
 	// Don't forget to remove the line we just processed from log.txt
 	delete_processed_line_from_log(text)!
 }
@@ -228,6 +240,25 @@ fn delete_processed_line_from_log(line string) ! {
 	os.write_file(log_txt, new_text)!
 }
 
+const db_strings = [
+	'db:',
+	'db.sqlite',
+	'db.mysql',
+	'db.pg',
+]
+
+const improvements_strings = [
+	'vfmt:',
+	'fmt:',
+	'all:',
+	'v:',
+]
+
+const parser_strings = [
+	'parser:',
+	'ast:',
+]
+
 const stdlib_strings = [
 	'gg:',
 	'json:',
@@ -244,33 +275,15 @@ const stdlib_strings = [
 	'toml:',
 	'vlib:',
 	'arrays:',
-]
+	'math.',
-
+	'os.',
-const db_strings = [
+	'term:',
-	'db:',
+	'sync.',
-	'db.sqlite',
+	'builtin:',
-	'db.mysql',
+	'strconv',
-]
+	'readline',
-
+	'cli:',
-const internal_strings = [
+	'eventbus:',
-	'scanner:',
-	'transformer:',
-]
-
-const improvements_strings = [
-	'vfmt:',
-	'fmt:',
-	'all:',
-]
-
-const tools_strings = [
-	'tools:',
-	'vpm:',
-]
-
-const parser_strings = [
-	'parser:',
-	'ast:',
 ]
 
 fn is_stdlib(text string) bool {
@@ -281,6 +294,22 @@ fn is_db(text string) bool {
 	return is_xxx(text, db_strings)
 }
 
+const orm_strings = [
+	'orm:',
+]
+
+fn is_orm(text string) bool {
+	return is_xxx(text, orm_strings)
+}
+
+const internal_strings = [
+	'scanner:',
+	'transformer:',
+	'builder:',
+	'pref:',
+	'v.util',
+]
+
 fn is_internal(text string) bool {
 	return is_xxx(text, internal_strings)
 }
@@ -289,6 +318,27 @@ fn is_improvements(text string) bool {
 	return is_xxx(text, improvements_strings)
 }
 
+const examples_strings = [
+	'example',
+	'tests',
+	'readme:',
+	'.md:',
+]
+
+fn is_examples(text string) bool {
+	return is_xxx(text, examples_strings)
+}
+
+const tools_strings = [
+	'tools:',
+	'vpm:',
+	'ci',
+	'github:',
+	'gitignore',
+	'benchmark',
+	'v.help:',
+]
+
 fn is_tools(text string) bool {
 	return is_xxx(text, tools_strings)
 }
@@ -297,6 +347,29 @@ fn is_parser(text string) bool {
 	return is_xxx(text, parser_strings)
 }
 
+const web_strings = [
+	'vweb:',
+	'websocket:',
+	'picoev:',
+	'mbedtls',
+	'net:',
+	'net.',
+	'wasm:',
+	'http:',
+]
+
+fn is_web(text string) bool {
+	return is_xxx(text, web_strings)
+}
+
+const native_strings = [
+	'native:',
+]
+
+fn is_native(text string) bool {
+	return is_xxx(text, native_strings)
+}
+
 fn is_xxx(text string, words []string) bool {
 	for s in words {
 		if text.contains(s) {