crate splitting: create fish-common crate

Dependencies between crates must form a DAG. This means that breaking up
the large library crate requires breaking dependency cycles. The goal of
this commit is creating a crate which contains some of the main crate's
functionality, without depending on the main crate.

To start off, we only move things required for extracting `src/wchar.rs`
and `src/wchar_ext.rs`, which will happen in a subsequent commit.

Part of #12182

Cargo.lock

@@ -160,6 +160,7 @@ dependencies = [
  "errno",
  "fish-build-helper",
  "fish-build-man-pages",
+ "fish-common",
  "fish-gettext",
  "fish-gettext-extraction",
  "fish-gettext-mo-file-parser",
@@ -200,6 +201,10 @@ dependencies = [
  "rsconf",
 ]
 
+[[package]]
+name = "fish-common"
+version = "0.0.0"
+
 [[package]]
 name = "fish-gettext"
 version = "0.0.0"

Cargo.toml

@@ -17,6 +17,7 @@ cfg-if = "1.0.3"
 errno = "0.3.0"
 fish-build-helper = { path = "crates/build-helper" }
 fish-build-man-pages = { path = "crates/build-man-pages" }
+fish-common = { path = "crates/common" }
 fish-gettext = { path = "crates/gettext" }
 fish-gettext-extraction = { path = "crates/gettext-extraction" }
 fish-gettext-maps = { path = "crates/gettext-maps" }
@@ -90,6 +91,7 @@ cfg-if.workspace = true
 errno.workspace = true
 fish-build-helper.workspace = true
 fish-build-man-pages = { workspace = true, optional = true }
+fish-common.workspace = true
 fish-gettext = { workspace = true, optional = true }
 fish-gettext-extraction = { workspace = true, optional = true }
 fish-printf.workspace = true

crates/common/Cargo.toml

@@ -0,0 +1,10 @@
+[package]
+name = "fish-common"
+edition.workspace = true
+rust-version.workspace = true
+version = "0.0.0"
+repository.workspace = true
+license.workspace = true
+
+[lints]
+workspace = true

crates/common/src/lib.rs

@@ -0,0 +1,28 @@
+// These are in the Unicode private-use range. We really shouldn't use this
+// range but have little choice in the matter given how our lexer/parser works.
+// We can't use non-characters for these two ranges because there are only 66 of
+// them and we need at least 256 + 64.
+//
+// If sizeof(wchar_t))==4 we could avoid using private-use chars; however, that
+// would result in fish having different behavior on machines with 16 versus 32
+// bit wchar_t. It's better that fish behave the same on both types of systems.
+//
+// Note: We don't use the highest 8 bit range (0xF800 - 0xF8FF) because we know
+// of at least one use of a codepoint in that range: the Apple symbol (0xF8FF)
+// on Mac OS X. See http://www.unicode.org/faq/private_use.html.
+pub const ENCODE_DIRECT_BASE: char = '\u{F600}';
+pub const ENCODE_DIRECT_END: char = char_offset(ENCODE_DIRECT_BASE, 256);
+
+pub const fn char_offset(base: char, offset: u32) -> char {
+    match char::from_u32(base as u32 + offset) {
+        Some(c) => c,
+        None => panic!("not a valid char"),
+    }
+}
+
+pub fn subslice_position<T: Eq>(a: &[T], b: &[T]) -> Option<usize> {
+    if b.is_empty() {
+        return Some(0);
+    }
+    a.windows(b.len()).position(|aw| aw == b)
+}

src/common.rs

@@ -17,6 +17,7 @@
 use crate::wildcard::{ANY_CHAR, ANY_STRING, ANY_STRING_RECURSIVE};
 use crate::wutil::fish_iswalnum;
 use bitflags::bitflags;
+use fish_common::{ENCODE_DIRECT_END, char_offset, subslice_position};
 use libc::{SIG_IGN, SIGTTOU, STDIN_FILENO};
 use once_cell::sync::OnceCell;
 use std::cell::{Cell, RefCell};
@@ -61,21 +62,6 @@
 // Unicode range for our needs.
 const _: () = assert!(WILDCARD_RESERVED_END <= RESERVED_CHAR_END);
 
-// These are in the Unicode private-use range. We really shouldn't use this
-// range but have little choice in the matter given how our lexer/parser works.
-// We can't use non-characters for these two ranges because there are only 66 of
-// them and we need at least 256 + 64.
-//
-// If sizeof(wchar_t))==4 we could avoid using private-use chars; however, that
-// would result in fish having different behavior on machines with 16 versus 32
-// bit wchar_t. It's better that fish behave the same on both types of systems.
-//
-// Note: We don't use the highest 8 bit range (0xF800 - 0xF8FF) because we know
-// of at least one use of a codepoint in that range: the Apple symbol (0xF8FF)
-// on Mac OS X. See http://www.unicode.org/faq/private_use.html.
-pub const ENCODE_DIRECT_BASE: char = '\u{F600}';
-pub const ENCODE_DIRECT_END: char = char_offset(ENCODE_DIRECT_BASE, 256);
-
 #[derive(Debug, Clone, Copy, PartialEq, Eq)]
 pub enum EscapeStringStyle {
     Script(EscapeFlags),
@@ -1016,13 +1002,6 @@ pub fn read_unquoted_escape(
     Some(in_pos)
 }
 
-pub const fn char_offset(base: char, offset: u32) -> char {
-    match char::from_u32(base as u32 + offset) {
-        Some(c) => c,
-        None => panic!("not a valid char"),
-    }
-}
-
 /// Exits without invoking destructors (via _exit), useful for code after fork.
 pub fn exit_without_destructors(code: libc::c_int) -> ! {
     unsafe { libc::_exit(code) };
@@ -1468,13 +1447,6 @@ fn slice_contains_slice<T: Eq>(a: &[T], b: &[T]) -> bool {
     subslice_position(a, b).is_some()
 }
 
-pub fn subslice_position<T: Eq>(a: &[T], b: &[T]) -> Option<usize> {
-    if b.is_empty() {
-        return Some(0);
-    }
-    a.windows(b.len()).position(|aw| aw == b)
-}
-
 #[derive(Copy, Debug, Clone, PartialEq, Eq)]
 pub enum WSL {
     Any,
@@ -2024,14 +1996,15 @@ macro_rules! env_stack_set_from_env {
 #[cfg(test)]
 mod tests {
     use super::{
-        ENCODE_DIRECT_BASE, ENCODE_DIRECT_END, ESCAPE_TEST_CHAR, EscapeFlags, EscapeStringStyle,
-        ScopeGuard, ScopedCell, ScopedRefCell, UnescapeStringStyle, bytes2wcstring, escape_string,
+        ENCODE_DIRECT_END, ESCAPE_TEST_CHAR, EscapeFlags, EscapeStringStyle, ScopeGuard,
+        ScopedCell, ScopedRefCell, UnescapeStringStyle, bytes2wcstring, escape_string,
         truncate_at_nul, unescape_string, wcs2bytes,
     };
     use crate::{
         util::get_seeded_rng,
         wchar::{L, WString, wstr},
     };
+    use fish_common::ENCODE_DIRECT_BASE;
     use rand::{Rng, RngCore};
 
     #[test]

src/env_universal_common.rs

@@ -806,11 +806,11 @@ fn skip_spaces(mut s: &wstr) -> &wstr {
 
 #[cfg(test)]
 mod tests {
+    use fish_common::ENCODE_DIRECT_BASE;
+    use fish_common::char_offset;
     use fish_tempfile::TempDir;
 
-    use crate::common::ENCODE_DIRECT_BASE;
     use crate::common::bytes2wcstring;
-    use crate::common::char_offset;
     use crate::common::wcs2osstring;
     use crate::env::{EnvVar, EnvVarFlags, VarTable};
     use crate::env_universal_common::{EnvUniversal, UvarFormat};

src/expand.rs

@@ -9,8 +9,8 @@
 };
 use crate::common::{
     EXPAND_RESERVED_BASE, EXPAND_RESERVED_END, EscapeFlags, EscapeStringStyle, UnescapeFlags,
-    UnescapeStringStyle, char_offset, charptr2wcstring, escape, escape_string,
-    escape_string_for_double_quotes, unescape_string, valid_var_name_char, wcs2zstring,
+    UnescapeStringStyle, charptr2wcstring, escape, escape_string, escape_string_for_double_quotes,
+    unescape_string, valid_var_name_char, wcs2zstring,
 };
 use crate::complete::{CompleteFlags, Completion, CompletionList, CompletionReceiver};
 use crate::env::{EnvVar, Environment};
@@ -30,6 +30,7 @@
 use crate::wildcard::{wildcard_expand_string, wildcard_has_internal};
 use crate::wutil::{Options, normalize_path, wcstoi_partial};
 use bitflags::bitflags;
+use fish_common::char_offset;
 use std::mem::MaybeUninit;
 
 bitflags! {

src/history/yaml_backend.rs

@@ -5,11 +5,10 @@
     time::{Duration, SystemTime, UNIX_EPOCH},
 };
 
+use fish_common::subslice_position;
+
 use super::{HistoryItem, PersistenceMode};
-use crate::{
-    common::{bytes2wcstring, subslice_position},
-    flog::flog,
-};
+use crate::{common::bytes2wcstring, flog::flog};
 
 // Our history format is nearly-valid YAML (but isn't quite). Here it is:
 //

src/wchar.rs

@@ -4,7 +4,7 @@
 //!   - wstr: a string slice without a nul terminator. Like `&str` but wide chars.
 //!   - WString: an owning string without a nul terminator. Like `String` but wide chars.
 
-use crate::common::{ENCODE_DIRECT_BASE, ENCODE_DIRECT_END};
+use fish_common::{ENCODE_DIRECT_BASE, ENCODE_DIRECT_END};
 pub use widestring::{Utf32Str as wstr, Utf32String as WString};
 
 /// Pull in our extensions.

src/wchar_ext.rs

@@ -2,9 +2,9 @@
 
 use crate::{
     L,
-    common::subslice_position,
     wchar::{WString, wstr},
 };
+use fish_common::subslice_position;
 use widestring::utfstr::CharsUtf32;
 
 /// Helpers to convert things to widestring.

src/wildcard.rs

@@ -1,5 +1,6 @@
 // Enumeration of all wildcard types.
 
+use fish_common::char_offset;
 use libc::X_OK;
 use once_cell::sync::Lazy;
 use std::cmp::Ordering;
@@ -7,7 +8,7 @@
 use std::os::unix::fs::MetadataExt;
 
 use crate::common::{
-    UnescapeFlags, UnescapeStringStyle, WILDCARD_RESERVED_BASE, WSL, char_offset,
+    UnescapeFlags, UnescapeStringStyle, WILDCARD_RESERVED_BASE, WSL,
     is_windows_subsystem_for_linux, unescape_string,
 };
 use crate::complete::{CompleteFlags, Completion, CompletionReceiver, PROG_COMPLETE_SEP};