diff options
| -rw-r--r-- | Cargo.lock | 293 | ||||
| -rw-r--r-- | Cargo.toml | 12 | ||||
| -rw-r--r-- | rust-toolchain.toml | 2 | ||||
| -rw-r--r-- | src/collect_files.rs | 494 | ||||
| -rw-r--r-- | src/config.rs | 92 | ||||
| -rw-r--r-- | src/generate_html.rs | 617 | ||||
| -rw-r--r-- | src/generate_rss.rs | 72 | ||||
| -rw-r--r-- | src/main.rs | 192 | ||||
| -rw-r--r-- | src/string_utils.rs | 145 |
9 files changed, 1345 insertions, 574 deletions
@@ -12,6 +12,21 @@ dependencies = [ ] [[package]] +name = "android_system_properties" +version = "0.1.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "819e7219dbd41043ac279b19830f2efc897156490d7fd6ea916720117ee66311" +dependencies = [ + "libc", +] + +[[package]] +name = "autocfg" +version = "1.5.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "c08606f8c3cbf4ce6ec8e28fb0014a2c086708fe954eaa885384a6165172e7e8" + +[[package]] name = "bit-set" version = "0.8.0" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -27,6 +42,47 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "5e764a1d40d510daf35e07be9eb06e75770908c27d411ee6c92109c9840eaaf7" [[package]] +name = "bumpalo" +version = "3.20.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "5d20789868f4b01b2f2caec9f5c4e0213b41e3e5702a50157d699ae31ced2fcb" + +[[package]] +name = "cc" +version = "1.2.56" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "aebf35691d1bfb0ac386a69bac2fde4dd276fb618cf8bf4f5318fe285e821bb2" +dependencies = [ + "find-msvc-tools", + "shlex", +] + +[[package]] +name = "cfg-if" +version = "1.0.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9330f8b2ff13f34540b44e946ef35111825727b38d33286ef986142615121801" + +[[package]] +name = "chrono" +version = "0.4.43" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "fac4744fb15ae8337dc853fee7fb3f4e48c0fbaa23d0afe49c447b4fab126118" +dependencies = [ + "iana-time-zone", + "js-sys", + "num-traits", + "wasm-bindgen", + "windows-link", +] + +[[package]] +name = "core-foundation-sys" +version = "0.8.7" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "773648b94d0e5d620f64f280777445740e61fe701025087ec8b57f45c791888b" + +[[package]] name = "fancy-regex" version = "0.14.0" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -38,14 +94,44 @@ dependencies = [ ] [[package]] +name = "find-msvc-tools" +version = "0.1.9" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "5baebc0774151f905a1a2cc41989300b1e6fbb29aff0ceffa1064fdd3088d582" + +[[package]] name = "highlight" -version = "1.0.0" -source = "git+git://benbridle.com/highlight?tag=v1.0.0#0e9fe32e7439a597a49f9c36c56a26726100b434" +version = "1.0.2" +source = "git+git://benbridle.com/highlight?tag=v1.0.2#e4802ab2a5422a87e4aadcb11b1cdb2d68f9c355" dependencies = [ "fancy-regex", ] [[package]] +name = "iana-time-zone" +version = "0.1.65" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e31bc9ad994ba00e440a8aa5c9ef0ec67d5cb5e5cb0cc7f8b744a35b389cc470" +dependencies = [ + "android_system_properties", + "core-foundation-sys", + "iana-time-zone-haiku", + "js-sys", + "log 0.4.29", + "wasm-bindgen", + "windows-core", +] + +[[package]] +name = "iana-time-zone-haiku" +version = "0.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "f31827a206f56af32e590ba56d5d2d085f558508192593743f16b2306495269f" +dependencies = [ + "cc", +] + +[[package]] name = "inked" version = "1.0.0" source = "git+git://benbridle.com/inked?tag=v1.0.0#2954d37b638fa2c1dd3d51ff53f08f475aea6ea3" @@ -54,6 +140,28 @@ dependencies = [ ] [[package]] +name = "js-sys" +version = "0.3.87" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "93f0862381daaec758576dcc22eb7bbf4d7efd67328553f3b45a412a51a3fb21" +dependencies = [ + "once_cell", + "wasm-bindgen", +] + +[[package]] +name = "libc" +version = "0.2.182" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "6800badb6cb2082ffd7b6a67e6125bb39f18782f793520caee8cb8846be06112" + +[[package]] +name = "log" +version = "0.4.29" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "5e5032e24019045c762d3c0f28f5b6b8bbf38563a65908389bf7978758920897" + +[[package]] name = "log" version = "1.1.1" source = "git+git://benbridle.com/log?tag=v1.1.1#930f3d0e2b82df1243f423c092a38546ea7533c3" @@ -68,8 +176,8 @@ dependencies = [ [[package]] name = "markdown" -version = "3.3.0" -source = "git+git://benbridle.com/markdown?tag=v3.3.0#df45ffb3affb7cb1d53b567b70fef721353ccffe" +version = "3.3.2" +source = "git+git://benbridle.com/markdown?tag=v3.3.2#e8d8a1146de5e297dda3dd6264a02f9d6938c3e1" [[package]] name = "memchr" @@ -78,12 +186,45 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "78ca9ab1a0babb1e7d5695e3530886289c18cf2f87ec19a575a0abdce112e3a3" [[package]] +name = "num-traits" +version = "0.2.19" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "071dfc062690e90b734c0b2273ce72ad0ffa95f0c74596bc250dcfd960262841" +dependencies = [ + "autocfg", +] + +[[package]] +name = "once_cell" +version = "1.21.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "42f5e15c9953c5e4ccceeb2e7382a716482c34515315f7b03532b8b4e8393d2d" + +[[package]] name = "paste" version = "1.0.15" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "57c0d7b74b563b49d38dae00a0c37d4d6de9b432382b2892f0574ddcae73fd0a" [[package]] +name = "proc-macro2" +version = "1.0.106" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "8fd00f0bb2e90d81d1044c2b32617f68fcb9fa3bb7640c23e9c748e53fb30934" +dependencies = [ + "unicode-ident", +] + +[[package]] +name = "quote" +version = "1.0.44" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "21b2ebcf727b7760c461f091f9f0f539b77b8e87f2fd88131e7f1b433b3cece4" +dependencies = [ + "proc-macro2", +] + +[[package]] name = "recipe" version = "1.4.0" source = "git+git://benbridle.com/recipe?tag=v1.4.0#652aaee3130e2ee02742fdcc248ddd1bee285737" @@ -106,15 +247,38 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "2b15c43186be67a4fd63bee50d0303afffcef381492ebe2c5d87f324e1b8815c" [[package]] +name = "rustversion" +version = "1.0.22" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "b39cdef0fa800fc44525c84ccb54a029961a8215f9619753635a9c0d2538d46d" + +[[package]] +name = "shlex" +version = "1.3.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0fda2ff0d084019ba4d7c6f371c95d8fd75ce3524c3cb8fb653a3023f6323e64" + +[[package]] name = "switchboard" -version = "1.0.0" -source = "git+git://benbridle.com/switchboard?tag=v1.0.0#ea70fa89659e5cf1a9d4ca6ea31fb67f7a2cc633" +version = "2.1.0" +source = "git+git://benbridle.com/switchboard?tag=v2.1.0#e6435712ba5b3ca36e99fc8cbe7755940f8b1f3f" dependencies = [ "log 1.1.1", "paste", ] [[package]] +name = "syn" +version = "2.0.117" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e665b8803e7b1d2a727f4023456bbbbe74da67099c585258af0ad9c5013b9b99" +dependencies = [ + "proc-macro2", + "quote", + "unicode-ident", +] + +[[package]] name = "termcolor" version = "1.4.1" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -125,8 +289,9 @@ dependencies = [ [[package]] name = "toaster" -version = "1.12.0" +version = "2.0.1" dependencies = [ + "chrono", "highlight", "log 2.0.0", "markdown", @@ -136,9 +301,60 @@ dependencies = [ ] [[package]] +name = "unicode-ident" +version = "1.0.24" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e6e4313cd5fcd3dad5cafa179702e2b244f760991f45397d14d4ebf38247da75" + +[[package]] name = "vagabond" -version = "1.1.1" -source = "git+git://benbridle.com/vagabond?tag=v1.1.1#b190582517e6008ad1deff1859f15988e4efaa26" +version = "1.1.2" +source = "git+git://benbridle.com/vagabond?tag=v1.1.2#5ace2626e7c8eba4647250346668052071107f0f" + +[[package]] +name = "wasm-bindgen" +version = "0.2.110" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "1de241cdc66a9d91bd84f097039eb140cdc6eec47e0cdbaf9d932a1dd6c35866" +dependencies = [ + "cfg-if", + "once_cell", + "rustversion", + "wasm-bindgen-macro", + "wasm-bindgen-shared", +] + +[[package]] +name = "wasm-bindgen-macro" +version = "0.2.110" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e12fdf6649048f2e3de6d7d5ff3ced779cdedee0e0baffd7dff5cdfa3abc8a52" +dependencies = [ + "quote", + "wasm-bindgen-macro-support", +] + +[[package]] +name = "wasm-bindgen-macro-support" +version = "0.2.110" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0e63d1795c565ac3462334c1e396fd46dbf481c40f51f5072c310717bc4fb309" +dependencies = [ + "bumpalo", + "proc-macro2", + "quote", + "syn", + "wasm-bindgen-shared", +] + +[[package]] +name = "wasm-bindgen-shared" +version = "0.2.110" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e9f9cdac23a5ce71f6bf9f8824898a501e511892791ea2a0c6b8568c68b9cb53" +dependencies = [ + "unicode-ident", +] [[package]] name = "winapi-util" @@ -150,6 +366,65 @@ dependencies = [ ] [[package]] +name = "windows-core" +version = "0.62.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "b8e83a14d34d0623b51dce9581199302a221863196a1dde71a7663a4c2be9deb" +dependencies = [ + "windows-implement", + "windows-interface", + "windows-link", + "windows-result", + "windows-strings", +] + +[[package]] +name = "windows-implement" +version = "0.60.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "053e2e040ab57b9dc951b72c264860db7eb3b0200ba345b4e4c3b14f67855ddf" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + +[[package]] +name = "windows-interface" +version = "0.59.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "3f316c4a2570ba26bbec722032c4099d8c8bc095efccdc15688708623367e358" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + +[[package]] +name = "windows-link" +version = "0.2.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "f0805222e57f7521d6a62e36fa9163bc891acd422f971defe97d64e70d0a4fe5" + +[[package]] +name = "windows-result" +version = "0.4.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "7781fa89eaf60850ac3d2da7af8e5242a5ea78d1a11c49bf2910bb5a73853eb5" +dependencies = [ + "windows-link", +] + +[[package]] +name = "windows-strings" +version = "0.5.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "7837d08f69c77cf6b07689544538e017c1bfcf57e34b4c0ff58e6c2cd3b37091" +dependencies = [ + "windows-link", +] + +[[package]] name = "windows-sys" version = "0.59.0" source = "registry+https://github.com/rust-lang/crates.io-index" @@ -1,15 +1,17 @@ [package] name = "toaster" -version = "1.12.0" +version = "2.0.1" edition = "2021" [dependencies] -vagabond = { git = "git://benbridle.com/vagabond", tag = "v1.1.1" } -markdown = { git = "git://benbridle.com/markdown", tag = "v3.3.0" } +vagabond = { git = "git://benbridle.com/vagabond", tag = "v1.1.2" } +markdown = { git = "git://benbridle.com/markdown", tag = "v3.3.2" } recipe = { git = "git://benbridle.com/recipe", tag = "v1.4.0" } log = { git = "git://benbridle.com/log", tag = "v2.0.0" } -switchboard = { git = "git://benbridle.com/switchboard", tag = "v1.0.0" } -highlight = { git = "git://benbridle.com/highlight", tag = "v1.0.0" } +switchboard = { git = "git://benbridle.com/switchboard", tag = "v2.1.0" } +highlight = { git = "git://benbridle.com/highlight", tag = "v1.0.2" } + +chrono = "0.4.43" [profile.release] lto=true diff --git a/rust-toolchain.toml b/rust-toolchain.toml new file mode 100644 index 0000000..5d56faf --- /dev/null +++ b/rust-toolchain.toml @@ -0,0 +1,2 @@ +[toolchain] +channel = "nightly" diff --git a/src/collect_files.rs b/src/collect_files.rs index ccdfc49..cf2f307 100644 --- a/src/collect_files.rs +++ b/src/collect_files.rs @@ -1,59 +1,75 @@ use crate::*; -use highlight::*; -use vagabond::*; - -use std::collections::HashMap; use std::fmt::Debug; pub struct Website { - pub name: String, - pub config: HashMap<String, String>, - pub highlighters: Highlighters, + pub name: Name, + pub config: Config, pub pages: Vec<Page>, pub redirects: Vec<Redirect>, - pub static_files: Vec<StaticItem>, // Redirects, !-prefixed-dir contents - pub static_dirs: Vec<StaticItem>, // Only !-prefixed static dirs + pub static_files: Vec<StaticFile>, // !-prefixed-dir contents + pub feeds: Vec<Feed>, // RSS feeds } pub struct Page { - pub name: String, // Display name of this page - pub name_url: String, // URL name for this page, no extension - pub full_url: String, // Full URL for this page, no extension - pub parents: Vec<String>, // Parent directory components, unsafe - pub parent_url: String, // Base URL for links in this page + pub name: Name, + pub url: String, // No extension + pub parents: Vec<Name>, // Parent names + pub parent_url: String, // With trailing slash pub source_path: PathBuf, // Absolute path to source file pub document: MarkdownDocument, // File content parsed as markdown pub headings: Vec<Heading>, // Ordered list of all headings in page - pub last_modified: Option<SystemTime>, // last-modified time of source file + pub last_modified: Option<SystemTime>, // Last-modified time of source file } pub struct Heading { - pub name: String, - pub url: String, + pub name: Name, + pub prefix: Option<Name>, // Disambiguation pub level: Level, + pub block_id: usize, // Pointer to heading element in document } -pub struct StaticItem { - pub full_url: String, // Safe full URL, with extension +impl Heading { + pub fn slug(&self) -> String { + match &self.prefix { + Some(prefix) => format!("{}-{}", prefix.slug(), self.name.slug()), + None => self.name.slug(), + } + } +} + +pub struct StaticFile { + pub url: String, // With extension pub source_path: PathBuf, // Absolute path to source file - pub last_modified: Option<SystemTime>, // last-modified time of source file + pub last_modified: Option<SystemTime>, // Last-modified time of source file +} + +pub struct Feed { + pub name: Name, // Taken from file name + pub url: String, // With extension + pub parents: Vec<Name>, // Parent names + pub parent_url: String, // Base URL for feed pages + pub source_path: PathBuf, // Absolute path to source file + pub last_modified: Option<SystemTime>, // Last-modified time of source file } pub struct Redirect { - pub name: String, // Display name of this redirect - pub full_url: String, // Safe full URL, no extension - pub parents: Vec<String>, // Parent directory components, unsafe + pub name: Name, + pub url: String, // No extension + pub parents: Vec<Name>, // Parent names pub parent_url: String, // Base URL for relative redirects - pub redirect: String, // Page to redirect to, as an internal link - pub last_modified: Option<SystemTime>, // last-modified time of source file + pub target: String, // Page to redirect to, internal link + pub source_path: PathBuf, // Absolute path to source file, for logging + pub last_modified: Option<SystemTime>, // Last-modified time of source file } + +/// Calculate correct relative path from this entity to a specified page. pub trait LinkFrom: Debug { - fn name(&self) -> &str; + fn name(&self) -> &Name; fn parent_url(&self) -> &str; - fn parents(&self) -> &[String]; + fn parents(&self) -> &[Name]; fn root(&self) -> String { let mut root = String::new(); for _ in self.parents() { @@ -67,11 +83,28 @@ pub trait LinkFrom: Debug { None => format!("/{}", self.name()), } } -} - -pub struct Highlighters { - pub languages: HashMap<String, usize>, - pub highlighters: Vec<Highlighter>, + /// Convert an internal link to a canonical page URL and optional heading, + /// both as slugs. + /// + /// `path` and returned URL have no extension. + fn canonicalise(&self, path: &str) -> (String, Option<String>) { + // Remove heading fragment from path. + let (path, heading) = match path.rsplit_once('#') { + Some((path, heading)) => match heading.is_empty() { + true => (path, None), + false => (path, Some(to_slug(heading))), + } + None => (path, None), + }; + let mut path = path.to_string(); + // Attach parent URL if not an absolute path. + if !path.starts_with('/') { + path = format!("{}{path}", self.parent_url()); + } + // Convert path to a canonical URL. + path = to_slug(&collapse_path(&path)); + return (path, heading); + } } impl Page { @@ -96,16 +129,33 @@ impl Debug for Redirect { } } +impl Debug for Feed { + fn fmt(&self, f: &mut std::fmt::Formatter) -> Result<(), std::fmt::Error> { + write!(f, "\"{}\"", self.qualified_name()) + } +} + impl LinkFrom for Page { - fn name(&self) -> &str { &self.name } + fn name(&self) -> &Name { &self.name } fn parent_url(&self) -> &str { &self.parent_url } - fn parents(&self) -> &[String] { &self.parents } + fn parents(&self) -> &[Name] { &self.parents } } impl LinkFrom for Redirect { - fn name(&self) -> &str { &self.name } + fn name(&self) -> &Name { &self.name } + fn parent_url(&self) -> &str { &self.parent_url } + fn parents(&self) -> &[Name] { &self.parents } +} + +impl LinkFrom for Feed { + fn name(&self) -> &Name { &self.name } fn parent_url(&self) -> &str { &self.parent_url } - fn parents(&self) -> &[String] { &self.parents } + fn parents(&self) -> &[Name] { &self.parents } +} + +pub struct ImagePaths { + pub thumb: String, + pub large: String, } @@ -115,43 +165,73 @@ impl Website { pages: Vec::new(), redirects: Vec::new(), static_files: Vec::new(), - static_dirs: Vec::new(), + feeds: Vec::new(), name: match Entry::from_path(path) { - Ok(entry) => entry.name, + Ok(entry) => entry.name.into(), Err(err) => fatal!("Couldn't open {:?}: {:?}", &path, err), }, - config: HashMap::new(), - highlighters: Highlighters { - languages: HashMap::new(), - highlighters: Vec::new(), - }, + config: Config::new(), }; + // Recursively collect entire website. new.collect_entry(path, path); - new.parse_highlighters(); + new.parse_hoisted_folders(); return new; } + /// Read the hoisted_folders config key, make root redirects for each + /// child of each listed directory. + fn parse_hoisted_folders(&mut self) { + for line in self.config.get("hoisted_folders").lines() { + if line.is_empty() { continue } + // Turn line into a path + let path = PathBuf::from(line); + let prefix: Vec<Name> = path.components() + .filter(|c| if let std::path::Component::Normal(_) = c {true} else {false}) + .map(|c| c.as_os_str().to_string_lossy().to_string()) + .map(|s| strip_numeric_prefix(&s).into()) + .collect(); + for page in &self.pages { + if page.parents == prefix { + let name = page.name.clone(); + let url = name.slug(); + let parents = Vec::new(); + let parent_url = String::new(); + let target = page.url.clone(); + let source_path = "<hoisted child>".into(); + let last_modified = self.config.last_modified; + self.redirects.push(Redirect { + name, + url, + parents, + parent_url, + target, + source_path, + last_modified, + }); + } + } + } + } + + /// Recursively collect an entry and all children. + /// `prefix` is the base directory path for the entire website. fn collect_entry(&mut self, path: &Path, prefix: &Path) { let entry = Entry::from_path(path).unwrap(); // Ignore dotted entries. if entry.name.starts_with('.') { return } - // Get name and extension. - let (mut name, extension) = entry.split_name(); - if let Some((prefix, suffix)) = name.split_once(' ') { - if prefix.chars().all(|c| "0123456789-".contains(c)) { - name = suffix.to_string(); - } - } - let name_url = make_url_safe(&name); - // Get last-modified time. + // Get name, extension, last-modified. + let (name_raw, extension) = entry.split_name(); + let name: Name = strip_numeric_prefix(&name_raw).into(); let last_modified = entry.last_modified; // Generate parent URL, used only for files. let source_path = entry.original_path.clone(); let relative_path = source_path.strip_prefix(prefix).unwrap_or_else( |_| fatal!("Path doesn't start with {prefix:?}: {source_path:?}")); - let mut parents: Vec<_> = relative_path.components() - .map(|c| c.as_os_str().to_string_lossy().to_string()).collect(); - parents.pop(); // Remove file segment. + let mut parents: Vec<Name> = relative_path.components() + .map(|c| c.as_os_str().to_string_lossy().to_string()) + .map(|s| strip_numeric_prefix(&s).into()) + .collect(); + parents.pop(); // Remove final (non-parent) segment. // Process each entry. if entry.is_directory() { @@ -162,11 +242,9 @@ impl Website { let relative_path = source_path.strip_prefix(&entry.original_path).unwrap_or_else( |_| fatal!("Path doesn't start with {prefix:?}: {source_path:?}")) .as_os_str().to_string_lossy().to_string(); - let full_url = format!("{stripped}/{relative_path}"); - self.static_files.push(StaticItem { full_url, source_path, last_modified }) + let url = format!("{stripped}/{relative_path}"); + self.static_files.push(StaticFile { url, source_path, last_modified }) } - let full_url = make_url_safe(stripped); - self.static_dirs.push(StaticItem { full_url, source_path, last_modified }); } else { for child in list_directory(entry.original_path).unwrap() { self.collect_entry(&child.original_path, prefix); @@ -174,57 +252,100 @@ impl Website { } } else if parents.is_empty() && entry.name.to_lowercase() == "toaster.conf" { info!("Reading configuration file at {path:?}"); - // Parse the config file. - let config = std::fs::read_to_string(&source_path).unwrap(); - let mut key = None; - let mut value = String::new(); - for line in config.lines() { - if line.starts_with(" ") || line.trim().is_empty() { - value.push_str(line.trim()); - value.push('\n'); - } else { - if let Some(key) = key { - self.config.insert(key, std::mem::take(&mut value)); - } - key = Some(line.trim().to_lowercase().to_string()); - } - } - if let Some(key) = key { - self.config.insert(key, std::mem::take(&mut value)); - } + let content = std::fs::read_to_string(&source_path).unwrap(); + self.config.parse_file(&content, last_modified); } else { - let full_name = match parents.last() { + // Used for error messages, to distinguish between pages of the same name. + let qualified_name = match parents.last() { Some(parent) => format!("{parent}/{name}"), None => name.to_string(), }; match extension.as_str() { + "feed" => { + let mut url = String::new(); + for parent in &parents { + url.push_str(&parent.slug()); + url.push('/'); + } + let parent_url = url.clone(); + url.push_str(&name.plain()); + self.feeds.push(Feed { + name, + url, + parents, + parent_url, + source_path, + last_modified, + }); + } + "redirect" => { + let mut url = String::new(); + for parent in &parents { + url.push_str(&parent.slug()); + url.push('/'); + } + let parent_url = url.clone(); + url.push_str(&name.slug()); + let target = std::fs::read_to_string(&source_path) + .unwrap().trim().to_string(); + self.redirects.push(Redirect { + name, + url, + parents, + parent_url, + target, + source_path, + last_modified, + }); + } "md" => { let markdown = std::fs::read_to_string(&source_path).unwrap(); let document = MarkdownDocument::from_str(&markdown); - let mut heading_set = HashSet::new(); - let mut duplicates = HashSet::new(); - let headings = document.blocks.iter() - .filter_map(|block| if let Block::Heading { line, level } = block { - let name = line.to_string(); - let url = make_url_safe(strip_appendix(&name)); + // Collect headings, check for duplicates. + let mut names_set = HashSet::new(); // all heading names + let mut duplicates = HashSet::new(); // only duplicates + let mut headings: Vec<Heading> = document.blocks.iter().enumerate() + .filter_map(|(block_id, block)| if let Block::Heading { line, level } = block { + let name: Name = line.to_string().into(); let level = level.to_owned(); - if !heading_set.insert(url.clone()) { - duplicates.insert(url.clone()); + let heading = Heading { name, prefix: None, level, block_id }; + if !names_set.insert(heading.slug()) { + duplicates.insert(heading.slug()); } - Some(Heading { name, url, level }) + Some(heading) } else { None }).collect(); - for url in duplicates { - warn!("Page {full_name:?} contains multiple headings with ID \"#{url}\""); + + // Namespace any duplicate headings to the parent h1 heading. + let mut prefix = None; + for heading in &mut headings { + if let Level::Heading1 = heading.level { + prefix = Some(heading.name.clone()); + } else { + if duplicates.contains(&heading.slug()) { + heading.prefix = prefix.clone(); + } + } + } + // Check for duplicates once more, and warn if any. + names_set.clear(); + duplicates.clear(); + for heading in &headings { + if !names_set.insert(heading.slug()) { + duplicates.insert(heading.slug()); + } } - if name_url == "+index" { + for slug in duplicates { + warn!("Page {qualified_name:?} contains multiple headings with ID \"#{slug}\""); + } + + if name.slug() == "+index" { if parents.is_empty() { // This is the index file for the whole site. self.pages.push(Page { - name: String::from("Home"), - name_url: String::from("index"), - full_url: String::from("index"), + name: self.name.clone(), + url: String::from("index"), parents, parent_url: String::from(""), source_path, @@ -234,20 +355,18 @@ impl Website { }); } else { // This is an index file for a directory. - let name = parents[parents.len()-1].clone(); - let name_url = make_url_safe(&name); - let mut full_url = String::new(); + let name = parents.last().unwrap().clone(); + let mut url = String::new(); for parent in &parents { - full_url.push_str(&make_url_safe(parent)); - full_url.push('/'); + url.push_str(&parent.slug()); + url.push('/'); } - let parent_url = full_url.clone(); - full_url.pop(); - parents.pop(); + let parent_url = url.clone(); + url.pop(); // Remove the trailing slash + parents.pop(); // Remove this directory self.pages.push(Page { name, - name_url, - full_url, + url, parents, parent_url, source_path, @@ -257,111 +376,72 @@ impl Website { }); } } else { - let mut full_url = String::new(); + // This is a regular page. + let mut url = String::new(); for parent in &parents { - full_url.push_str(&make_url_safe(parent)); - full_url.push('/'); + url.push_str(&parent.slug()); + url.push('/'); } - full_url.push_str(&name_url); - let mut parent_url = full_url.clone(); + // Children descend from this page, so the parent + // url must contain this page. + url.push_str(&name.slug()); + let mut parent_url = url.clone(); parent_url.push('/'); self.pages.push(Page { - name, name_url, full_url, - parents, parent_url, + name, + url, + parents, + parent_url, source_path, - document, headings, + document, + headings, last_modified, }); } }, - "redirect" => { - let mut full_url = String::new(); - for parent in &parents { - full_url.push_str(&make_url_safe(parent)); - full_url.push('/'); - } - let parent_url = full_url.clone(); - full_url.push_str(&name_url); - let redirect = std::fs::read_to_string(&source_path) - .unwrap().trim().to_string(); - self.redirects.push(Redirect { - name, full_url, parents, parent_url, - redirect, last_modified, - }); - } _ => { + // This is a static file. let mut parent_url = String::new(); for parent in &parents { - parent_url.push_str(&make_url_safe(parent)); - parent_url.push('/'); + parent_url.push_str(&parent.slug()); + parent_url.push('/'); } - let full_url = format!("{parent_url}{name_url}.{extension}"); - self.static_files.push(StaticItem { full_url, source_path, last_modified }); + let name_slug = name.slug(); + let url = format!("{parent_url}{name_slug}.{extension}"); + self.static_files.push(StaticFile { url, source_path, last_modified }); }, } } } - pub fn parse_highlighters(&mut self) { - let mut languages = Vec::new(); - let mut source = String::new(); - for line in self.get_config("highlighters").lines() { - if let Some(line) = line.trim().strip_prefix('[') { - if let Some(line) = line.strip_suffix(']') { - // Bank the current source. - if !languages.is_empty() { - let i = self.highlighters.highlighters.len(); - for language in languages { - self.highlighters.languages.insert(language, i); - } - let highlighter = Highlighter::from_str(&source); - self.highlighters.highlighters.push(highlighter); - } - languages = line.split('/').map(|s| s.trim().to_string()).collect(); - source.clear(); - continue; - } - } - source.push_str(line); - source.push('\n'); - } - // Bank the current source. - if !languages.is_empty() { - let i = self.highlighters.highlighters.len(); - for language in languages { - self.highlighters.languages.insert(language, i); - } - let highlighter = Highlighter::from_str(&source); - self.highlighters.highlighters.push(highlighter); - } - } - - // Ext is extension without a dot. - // Checks if a relative link to an internal page name can be reached from - // the current page, and returns a resolved absolute link to the page with extension. + /// Check if the internal link `path` is valid, pointing to a real internal + /// page with extension `ext` and heading, relative to the current page (`from`). + /// Returns a resolved absolute link to the page, with extension. pub fn has_page(&self, from: &impl LinkFrom, path: &str, ext: &str) -> Option<String> { - // Remove heading fragment and file extension. - let (path, heading) = match path.rsplit_once('#') { - Some((path, heading)) => match heading.is_empty() { - true => (path, None), - false => (path, Some(heading)), - } - None => (path, None), + let original_path = path; + let (mut path, mut heading) = from.canonicalise(path); + if let Some(stripped) = path.strip_suffix(&format!(".{ext}")) { + path = stripped.to_string(); }; - let mut path = path.strip_suffix(&format!(".{ext}")).unwrap_or(path).to_string(); - // Attach parent if not an absolute path. - if !path.starts_with('/') { - path = format!("{}{path}", from.parent_url()); - } - let path = make_url_safe(&collapse_path(&path)); - // Find page with this path in website. + // Find page with this path in website, resolving any redirect first. + for redirect in &self.redirects { + if redirect.url == path { + let (target_path, target_heading) = redirect.canonicalise(&redirect.target); + path = target_path; + if target_heading.is_some() && heading.is_some() { + warn!("Page {from:?} contains link {original_path:?} to a redirect that also links to a heading"); + } + if heading.is_none() { + heading = target_heading; + } + } + } for page in &self.pages { - if page.full_url == path { + if page.url == path { let root = from.root(); if let Some(heading) = heading { - let heading = make_url_safe(strip_appendix(heading)); - if !page.headings.iter().any(|h| h.url == heading) { + if !page.headings.iter().any(|h| h.slug() == heading) { warn!("Page {from:?} contains link to nonexistent heading {heading:?} on page {path:?}"); } return Some(format!("{root}{path}.{ext}#{heading}")); @@ -373,28 +453,46 @@ impl Website { return None; } + /// Check if the external link `path` points to a valid static file or + /// generated feed. Returns a resolved absolute link to the file. pub fn has_static(&self, from: &impl LinkFrom, path: &str) -> Option<String> { // Attach parent if not an absolute path. + // We don't want to canonicalise/sluggify the path. let path = match !path.starts_with('/') { true => collapse_path(&format!("{}{path}", from.parent_url())), false => collapse_path(path), }; + let root = from.root(); for file in &self.static_files { - if file.full_url == path { - let root = from.root(); + if file.url == path { + return Some(format!("{root}{path}")); + } + } + for feed in &self.feeds { + if format!("{}.rss", feed.url) == path { return Some(format!("{root}{path}")); } } return None; } - pub fn has_image(&self, file_name: &str) -> bool { - let image_path = format!("images/thumb/{file_name}"); - self.static_files.iter().any(|s| s.full_url == image_path) - } + /// Check if a particular image exists. + pub fn has_image(&self, file_name: &str, root: &str) -> Option<ImagePaths> { + let check = |path: String| + match self.static_files.iter().any(|s| s.url == path) { + true => Some(format!("{root}{path}")), + false => None, + }; + let thumb_path = check(format!("images/thumb/{file_name}")); + let large_path = check(format!("images/large/{file_name}")); + let fallback_path = check(format!("images/{file_name}")) + .or(large_path.clone()) + .or(thumb_path.clone()); - pub fn get_config(&self, key: &str) -> String { - self.config.get(key).map(String::to_owned).unwrap_or_else(String::new) + Some(ImagePaths { + thumb: thumb_path.or(fallback_path.clone())?, + large: large_path.or(fallback_path.clone())?, + }) } } @@ -419,3 +517,13 @@ fn collapse_path(path: &str) -> String { return segments.join("/"); } } + + +fn strip_numeric_prefix(name: &str) -> String { + if let Some((prefix, suffix)) = name.split_once(' ') { + if prefix.chars().all(|c| "0123456789-".contains(c)) { + return suffix.to_string(); + } + } + return name.to_string(); +} diff --git a/src/config.rs b/src/config.rs new file mode 100644 index 0000000..1933a70 --- /dev/null +++ b/src/config.rs @@ -0,0 +1,92 @@ +use crate::*; + +use highlight::*; + +use std::collections::HashMap; + + +pub struct Config { + pub config: HashMap<String, String>, + pub last_modified: Option<SystemTime>, + pub languages: HashMap<String, usize>, + pub highlighters: Vec<Highlighter>, + pub root_redirects: Vec<String>, +} + +impl Config { + pub fn new() -> Self { + Self { + config: HashMap::new(), + last_modified: None, + languages: HashMap::new(), + highlighters: Vec::new(), + root_redirects: Vec::new(), + } + } + + pub fn get(&self, key: &str) -> String { + match self.config.get(key) { + Some(value) => value.to_owned(), + None => String::new(), + } + } + + pub fn parse_file(&mut self, content: &str, last_modified: Option<SystemTime>) { + if self.last_modified.is_none() || self.last_modified > last_modified { + self.last_modified = last_modified; + } + + let mut key = None; + let mut value = String::new(); + macro_rules! bank_value { + () => { + value = value.trim().to_string(); + if let Some(key) = key { + if key == "highlighters" { + self.parse_highlighters(&value); + } + self.config.insert(key, std::mem::take(&mut value)); + } + }; + } + for line in content.lines() { + if line.starts_with(" ") || line.trim().is_empty() { + value.push_str(line.trim()); + value.push('\n'); + } else { + bank_value!(); + key = Some(line.trim().to_lowercase().to_string()); + } + } + bank_value!(); + } + + fn parse_highlighters(&mut self, value: &str) { + let mut languages = Vec::new(); + let mut source = String::new(); + macro_rules! bank_sources { + () => { + if !languages.is_empty() { + let i = self.highlighters.len(); + for language in std::mem::take(&mut languages) { + self.languages.insert(language, i); + } + let highlighter = Highlighter::from_str(&std::mem::take(&mut source)); + self.highlighters.push(highlighter); + } + }; + } + for line in value.lines() { + if let Some(line) = line.trim().strip_prefix('[') { + if let Some(line) = line.strip_suffix(']') { + bank_sources!(); + languages = line.split('/').map(|s| s.trim().to_string()).collect(); + continue; + } + } + source.push_str(line); + source.push('\n'); + } + bank_sources!(); + } +} diff --git a/src/generate_html.rs b/src/generate_html.rs index af48d2e..3b6ee62 100644 --- a/src/generate_html.rs +++ b/src/generate_html.rs @@ -4,61 +4,98 @@ use markdown::*; use recipe::*; -pub fn generate_html(document: &MarkdownDocument, page: &Page, website: &Website) -> String { +const DEFAULT_TEMPLATE: &str = "\ +<!DOCTYPE html> + <head> + <title>{page_name} — {site_name}</title> + <meta charset='UTF-8'> + <meta name='viewport' content='width=device-width, initial-scale=1'> + {head} + </head> + <body> + <header> + <nav id='up'> + {home_link} + {parent_link} + </nav> + <h1 id='title'>{page_name_html}</h1> + </header> + <main> + {main} + </main> + </body> +</html>"; + + +pub fn generate_html(page: &Page, website: &Website) -> String { let root = page.root(); - let page_name = sanitize_text(&page.name, true); - let site_name = sanitize_text(&website.name, true); + + // Get page name as a plain string and as an HTML fragment. + let mut page_name = page.name.plain(); + let mut page_name_html = sanitize_text(&page_name, true); + // Find any override-title fragments. + for block in &page.document.blocks { + if let Block::Fragment { language, content } = block { + if language == "override-title" { + let line = Line::from_str(content); + page_name = line.to_string(); + page_name_html = line_to_html(&line, page, website, &None); + } + } + } + page_name = sanitize_text(&page_name, true); + + // Get the URL of the parent page. + let site_name = sanitize_text(&website.name.plain(), true); let mut parent_url = String::new(); for segment in &page.parents { - parent_url.push_str(&make_url_safe(segment)); parent_url.push('/'); + parent_url.push_str(&segment.slug()); parent_url.push('/'); } parent_url.pop(); let head = get_html_head(page, website); let head = head.trim(); let home_link = format!("<a id='home' href='{root}index.html'>{site_name}</a>"); let parent_link = match page.parents.last() { - Some(name) => format!("<a id='parent' href='../{}.html'>{name}</a>", make_url_safe(name)), + Some(name) => format!("<a id='parent' href='../{}.html'>{name}</a>", name.slug()), None => String::new(), }; - let table_of_contents = get_table_of_contents(page); - let main = document_to_html(document, page, website); let main = main.trim(); - format!("\ -<!DOCTYPE html> -<head> -<title>{page_name} — {site_name}</title> -<meta charset='UTF-8'> -<meta name='viewport' content='width=device-width, initial-scale=1'> -{head} -</head> -<body> -<header> -<nav id='up'> -{home_link} -{parent_link} -</nav> -<h1 id='title'>{page_name}</h1> -<nav id='toc'> -{table_of_contents} -</nav> -</header> -<main> -{main} -</main> -</body> -</html>") + // Format tables of contents and the main page. + let toc = get_table_of_contents(page); + let toc_compact = if page.headings.len() >= 3 { + format!("<details><summary></summary>\n{toc}</details>\n") + } else { String::new() }; + let main = document_to_html(page, website); let main = main.trim(); + + let mut template = website.config.get("html.template"); + if template.trim().is_empty() { + template = DEFAULT_TEMPLATE.to_string(); + } + + template + .replace("{site_name}", &site_name ) + .replace("{page_name}", &page_name ) + .replace("{page_name_html}", &page_name_html) + .replace("{home_link}", &home_link ) + .replace("{parent_link}", &parent_link ) + .replace("{head}", &head ) + .replace("{toc_compact}", &toc_compact ) + .replace("{toc}", &toc ) + .replace("{main}", &main ) } -pub fn generate_html_redirect(path: &str) -> String { +pub fn generate_html_redirect(path: &str, website: &Website) -> String { + let head = website.config.get("html.redirect.head"); let head = head.trim(); let path = sanitize_text(path, false); format!("\ <!DOCTYPE html> <head> -<title>Redirect</title> +<title>Redirecting...</title> <meta http-equiv='refresh' content='0; url={path}'> +{head} </head> -<html>") +</html>") } @@ -77,7 +114,7 @@ pub fn get_html_head(page: &Page, website: &Website) -> String { } } if include_default_head { - html_head.insert_str(0, &website.get_config("html.head")); + html_head.insert_str(0, &website.config.get("html.head")); } let root = page.root(); html_head @@ -87,16 +124,13 @@ pub fn get_html_head(page: &Page, website: &Website) -> String { pub fn get_table_of_contents(page: &Page) -> String { - if page.headings.len() < 3 { - return String::new(); - } - let mut toc = String::from("<details><summary></summary><ul>\n"); - let site_name = sanitize_text(&page.name, true); - toc.push_str(&format!("<li class='l1'><a href='#title'>{site_name}</a></li>\n")); + let mut toc = String::from("<ul>\n"); + let page_name = sanitize_text(&page.name.plain(), true); + toc.push_str(&format!("<li class='l1'><a href='#title'>{page_name}</a></li>\n")); for heading in &page.headings { - let name = &heading.name; - let url = &heading.url; + let name = sanitize_text(&heading.name.plain(), true); + let url = &heading.slug(); let class = match heading.level { Level::Heading1 => "l1", Level::Heading2 => "l2", @@ -104,17 +138,19 @@ pub fn get_table_of_contents(page: &Page) -> String { }; toc.push_str(&format!("<li class='{class}'><a href='#{url}'>{name}</a></li>\n")); } - toc.push_str("</ul></details>\n"); + toc.push_str("</ul>\n"); return toc; } -pub fn document_to_html(document: &MarkdownDocument, page: &Page, website: &Website) -> String { +pub fn document_to_html(page: &Page, website: &Website) -> String { let from = &page; + let root = page.root(); let mut html = String::new(); + let mut prefix = None; macro_rules! line_to_html { - ($l:expr) => {{ line_to_html(&$l, page, website) }}; } + ($l:expr) => {{ line_to_html(&$l, page, website, &prefix) }}; } macro_rules! html { ($($arg:tt)*) => {{ html.push_str(&format!($($arg)*)); html.push('\n'); }}; } macro_rules! tag { @@ -124,201 +160,236 @@ pub fn document_to_html(document: &MarkdownDocument, page: &Page, website: &Webs ($t:expr,$c:expr,$f:expr) => {{ html!("<{} {}>", $t, $c); $f; html!("</{}>", $t); }}; ($t:expr,$f:expr) => {{ html!("<{}>", $t); $f; html!("</{}>", $t); }}; } - let root = page.root(); - for block in &document.blocks { - match block { - Block::Heading { level, line } => { - let id = make_url_safe(strip_appendix(&line.to_string())); - match level { - Level::Heading1 => tag!("h1", line, format!("id='{id}'")), - Level::Heading2 => tag!("h2", line, format!("id='{id}'")), - Level::Heading3 => tag!("h3", line, format!("id='{id}'")), - } - } - Block::Paragraph(line) => tag!("p", line), - Block::Math(content) => html!("<div class='math'>{}</div>", sanitize_text(content, false)), - Block::List(lines) => wrap!("ul", for line in lines { - // Insert a <br> tag directly after the first untagged colon. - let mut depth = 0; - let mut prev = '\0'; - let mut output = String::new(); - let mut class = String::new(); - for c in line_to_html!(line).chars() { - output.push(c); - if c == '<' { - depth += 1; - } else if c == '/' && prev == '<' { - depth -= 2; // 2 because prev was a '<' as well. - } else if c == ':' && depth == 0 { - output.pop(); output.push_str("<br>"); - class.push_str("extended"); depth += 99; + wrap!("article", + for (i, block) in page.document.blocks.iter().enumerate() { + match block { + Block::Heading { level, line } => { + if let Level::Heading1 = level { + html!("</article>"); + html!("<article>"); + prefix = Some(to_slug(&line.to_string())); + }; + // Find namespaced heading ID from headings list. + let url = match page.headings.iter().find(|h| h.block_id == i) { + Some(heading) => heading.slug(), + None => unreachable!("Couldn't find heading in headings list"), + }; + let heading_tag = match level { + Level::Heading1 => "h1", + Level::Heading2 => "h2", + Level::Heading3 => "h3", + }; + // Find out whether line contains a link. + let mut contains_link = false; + for token in &line.tokens { + if let Token::InternalLink { .. } | Token::ExternalLink { .. } = token { + contains_link = true; + break; + } } - prev = c; - } - // Replace a leading checkbox with a real checkbox. - if let Some(stripped) = output.strip_prefix("<code>[ ]</code>") { - output = format!("<input type='checkbox' disabled>{stripped}"); - class.push_str(" checkbox"); - } else if let Some(stripped) = output.strip_prefix("<code>[x]</code>") { - output = format!("<input type='checkbox' disabled checked>{stripped}"); - class.push_str(" checkbox"); - } else if let Some(stripped) = output.strip_prefix("<code>[X]</code>") { - output = format!("<input type='checkbox' disabled checked>{stripped}"); - class.push_str(" checkbox"); - }else if let Some(stripped) = output.strip_prefix("[ ]") { - output = format!("<input type='checkbox' disabled>{stripped}"); - class.push_str(" checkbox"); - } else if let Some(stripped) = output.strip_prefix("[x]") { - output = format!("<input type='checkbox' disabled checked>{stripped}"); - class.push_str(" checkbox"); - } else if let Some(stripped) = output.strip_prefix("[X]") { - output = format!("<input type='checkbox' disabled checked>{stripped}"); - class.push_str(" checkbox"); - } - let class = class.trim(); - match class.is_empty() { - true => html!("<li>{output}</li>"), - false => html!("<li class='{class}'>{output}</li>"), + if !contains_link { + wrap!(heading_tag, format!("id='{url}'"), { + tag!("a", line, format!("href='#{url}'")); + }); + } else { + // Leave off the heading <a> tag if the heading already contains + // a link (<a> tags cannot be nested). + tag!(heading_tag, line, format!("id='{url}'")); + } + } - }), - Block::Note(lines) => wrap!("aside", for line in lines { tag!("p", line) }), - Block::Embed { label, path } => match path.rsplit_once('.') { - Some((_, extension)) => { - let mut path = path.to_string(); - if !path.contains("://") { - match website.has_static(page, &path) { - Some(resolved) => path = resolved, - None => warn!("Page {from:?} embeds nonexistent static file {path:?}"), + Block::Paragraph(line) => tag!("p", line), + Block::Math(content) => html!("<div class='math'>{}</div>", sanitize_text(content, false)), + Block::List(lines) => wrap!("ul", for line in lines { + // Insert a <br> tag directly after the first untagged colon. + let mut depth = 0; + let mut prev = '\0'; + let mut output = String::new(); + let mut class = String::new(); + for c in line_to_html!(line).chars() { + output.push(c); + if c == '<' { + depth += 1; + } else if c == '/' && prev == '<' { + depth -= 2; // 2 because prev was a '<' as well. + } else if c == ' ' && prev == ':' && depth == 0 { + output.pop(); output.pop(); output.push_str("<br>"); + class.push_str("extended"); depth += 99; } + prev = c; } - let label = sanitize_text(label, true); - let path = sanitize_text(&path, false); - match extension.to_lowercase().as_str() { - "jpg"|"jpeg"|"png"|"webp"|"gif"|"tiff" => html!( - "<figure><a href='{path}'><img src='{path}' alt='{label}' title='{label}' /></a></figure>"), - "mp3"|"wav"|"m4a" => html!("<audio controls src='{path}'>{label}</audio>"), - "mp4"|"avi" => html!("<video controls src='{path}'>{label}</video>"), - ext @ _ => warn!("Unrecognised extension for embedded file {path:?} with extension {ext:?} in page {from:?}"), + // Replace a leading checkbox with a real checkbox. + if let Some(stripped) = output.strip_prefix("<code>[ ]</code>") { + output = format!("<input type='checkbox' disabled>{stripped}"); + class.push_str(" checkbox"); + } else if let Some(stripped) = output.strip_prefix("<code>[x]</code>") { + output = format!("<input type='checkbox' disabled checked>{stripped}"); + class.push_str(" checkbox"); + } else if let Some(stripped) = output.strip_prefix("<code>[X]</code>") { + output = format!("<input type='checkbox' disabled checked>{stripped}"); + class.push_str(" checkbox"); + }else if let Some(stripped) = output.strip_prefix("[ ]") { + output = format!("<input type='checkbox' disabled>{stripped}"); + class.push_str(" checkbox"); + } else if let Some(stripped) = output.strip_prefix("[x]") { + output = format!("<input type='checkbox' disabled checked>{stripped}"); + class.push_str(" checkbox"); + } else if let Some(stripped) = output.strip_prefix("[X]") { + output = format!("<input type='checkbox' disabled checked>{stripped}"); + class.push_str(" checkbox"); } - } - _ => warn!("Cannot embed file {path:?} with no file extension in page {from:?}"), - } - Block::Fragment { language, content } => { - match language.as_str() { - "math" => html!("<div class='math'>{}</div>", content.replace("\n", " \\\\\n")), - "embed-html" => html!("{content}"), - "embed-css" => wrap!("style", html!("{content}")), - "embed-javascript"|"embed-js" => wrap!("script", html!("{content}")), - "embed-html-head"|"override-html-head" => (), - "hidden"|"todo" => (), - "poem" => wrap!("div", "class='poem'", for line in content.lines() { - let line = line.trim_end(); - match line.is_empty() { - true => html!("<br>"), - false => html!("<p>{}</p>", sanitize_text(line, true)), + let class = class.trim(); + match class.is_empty() { + true => html!("<li>{output}</li>"), + false => html!("<li class='{class}'>{output}</li>"), + } + }), + Block::Note(lines) => wrap!("aside", for line in lines { tag!("p", line) }), + Block::Embed { label, path } => match path.rsplit_once('.') { + Some((_, extension)) => { + let mut path = path.to_string(); + if !path.contains("://") { + match website.has_static(page, &path) { + Some(resolved) => path = resolved, + None => warn!("Page {from:?} embeds nonexistent static file {path:?}"), + } } - }), - "recipe" => { - let recipe = Recipe::parse(content); - html!("<div class='recipe'><ul>"); - for ingredient in recipe.ingredients { html!("<li>{ingredient}</li>") } - html!("</ul><hr>"); - for paragraph in recipe.process { html!("<p>{paragraph}</p>") } - html!("</div>"); - }, - "gallery" => wrap!("div", "class='gallery'", for line in content.lines() { - let file = line.trim(); - if !website.has_image(file) { - warn!("Gallery on page {from:?} references nonexistent image {file:?}"); - continue; + let label = sanitize_text(label, true); + let path = sanitize_text(&path, false); + match extension.to_lowercase().as_str() { + "jpg"|"jpeg"|"png"|"webp"|"gif"|"tiff" => html!( + "<figure><a href='{path}'><img src='{path}' alt='{label}' title='{label}' /></a></figure>"), + "mp3"|"wav"|"m4a" => html!("<audio controls src='{path}'>{label}</audio>"), + "mp4"|"avi" => html!("<figure><video controls src='{path}'>{label}</video></figure>"), + ext @ _ => warn!("Unrecognised extension for embedded file {path:?} with extension {ext:?} in page {from:?}"), } - let large = sanitize_text(&format!("{root}images/large/{file}"), false); - // let small = sanitize_text(&format!("{root}images/small/{file}"), false); - let thumb = sanitize_text(&format!("{root}images/thumb/{file}"), false); - html!("<a href='{large}'><img src='{thumb}' /></a>"); - }), - "gallery-nav" => wrap!("div", "class='gallery-nav'", for line in content.lines() { - let line = line.trim(); - if let Some((name, image)) = line.split_once("::") { - let name = name.trim(); - let image = image.trim(); - let ParsedLink { path, class, label } = parse_internal_link(name, page, website); - if website.has_image(image) { - let thumb = sanitize_text(&format!("{root}images/thumb/{image}"), false); - html!("<a href='{path}' class='{class}'><img src='{thumb}'/><p>{label}</p></a>") + } + _ => warn!("Cannot embed file {path:?} with no file extension in page {from:?}"), + } + Block::Fragment { language, content } => { + match language.as_str() { + "math" => { + let mut content = content.trim().to_string(); + if !content.contains(r"\begin") { + content = content.replace("\n", " \\\\\n").to_string(); + } + html!("<div class='math'>{content}</div>") + }, + "embed-html" => html!("{content}"), + "embed-css" => wrap!("style", html!("{content}")), + "embed-javascript"|"embed-js" => wrap!("script", html!("{content}")), + "embed-html-head"|"override-html-head"|"override-title" => (), + "hidden"|"todo" => (), + "poem" => wrap!("div", "class='poem'", for line in content.lines() { + let line = line.trim_end(); + match line.is_empty() { + true => html!("<br>"), + false => html!("<p>{}</p>", sanitize_text(line, true)), + } + }), + "recipe" => { + let recipe = Recipe::parse(content); + html!("<div class='recipe'><ul>"); + for ingredient in recipe.ingredients { html!("<li>{ingredient}</li>") } + html!("</ul><hr>"); + for paragraph in recipe.process { html!("<p>{paragraph}</p>") } + html!("</div>"); + }, + "gallery" => wrap!("div", "class='gallery'", for line in content.lines() { + let file_name = line.trim(); + if let Some(image_paths) = website.has_image(file_name, &root) { + let large = sanitize_text(&image_paths.large, false); + let thumb = sanitize_text(&image_paths.thumb, false); + html!("<a href='{large}'><img src='{thumb}' /></a>"); } else { - warn!("Gallery-nav on page {from:?} references nonexistent image {image:?}"); + warn!("Gallery on page {from:?} references nonexistent image {file_name:?}"); } - } else { - warn!("Gallery-nav on page {from:?} has line without a '::' separator"); - } - }), - _ => wrap!("pre", format!("class='{language}'"), { - if let Some(i) = website.highlighters.languages.get(language) { - let mut source = String::new(); - let highlighter = &website.highlighters.highlighters[*i]; - for span in highlighter.highlight(content) { - if span.tag.is_empty() { - source.push_str(&sanitize_text(&span.text, false)); + }), + "gallery-nav" => wrap!("div", "class='gallery-nav'", for line in content.lines() { + let line = line.trim(); + if let Some((name, file_name)) = line.split_once("::") { + let name = name.trim(); + let file_name = file_name.trim(); + let ParsedLink { path, class, label } = parse_internal_link(name, page, website, &prefix); + if let Some(image_paths) = website.has_image(file_name, &root) { + let thumb = sanitize_text(&image_paths.thumb, false); + html!("<a href='{path}' class='{class}'><img src='{thumb}'/><p>{label}</p></a>") } else { - source.push_str(&format!("<span class='{}'>", span.tag.to_lowercase())); - source.push_str(&sanitize_text(&span.text, false)); - source.push_str("</span>"); + warn!("Gallery on page {from:?} references nonexistent image {file_name:?}"); + warn!("Gallery-nav on page {from:?} references nonexistent image {file_name:?}"); } + } else { + warn!("Gallery-nav on page {from:?} has line without a '::' separator"); } - html!("{source}"); - } else { - html!("{}", sanitize_text(content, false)) - } - }) + }), + _ => wrap!("pre", format!("class='{language}'"), { + if let Some(i) = website.config.languages.get(language) { + let mut source = String::new(); + let highlighter = &website.config.highlighters[*i]; + for span in highlighter.highlight(content) { + if span.tag.is_empty() { + source.push_str(&sanitize_text(&span.text, false)); + } else { + source.push_str(&format!("<span class='{}'>", span.tag.to_lowercase())); + source.push_str(&sanitize_text(&span.text, false)); + source.push_str("</span>"); + } + } + html!("{source}"); + } else { + html!("{}", sanitize_text(content, false)) + } + }) + } } - } - Block::Break => html!("<hr>"), - Block::Table(table) => wrap!("div", "class='table'", wrap!("table", { - wrap!("thead", - wrap!("tr", for column in &table.columns { - match column.border_right { - true => tag!("th", column.name, "class='border'"), - false => tag!("th", column.name), - } - }) - ); - for section in &table.sections { - wrap!("tbody", for row in section { - wrap!("tr", for (column, cell) in std::iter::zip(&table.columns, row) { - let text_raw = line_to_html!(cell); - let text = match text_raw.as_str() { - "Yes" => "✓", - "No" => "✗", - other => other, - }; - let mut class = match text { - "--" => "c", - _ => match column.alignment { - Alignment::Left => "l", - Alignment::Center => "c", - Alignment::Right => "r", - }, - }.to_string(); - if ["No", "--", "0"].contains(&text_raw.as_str()) { - class.push_str(" dim"); - }; - if column.border_right { - class.push_str(" border"); + Block::Break => html!("<hr>"), + Block::Table(table) => wrap!("div", "class='table'", wrap!("table", { + wrap!("thead", + wrap!("tr", for column in &table.columns { + match column.border_right { + true => tag!("th", column.name, "class='border'"), + false => tag!("th", column.name), } - html!("<td class='{class}'>{text}</td>"); }) - }) - }; - })) + ); + for section in &table.sections { + wrap!("tbody", for row in section { + wrap!("tr", for (column, cell) in std::iter::zip(&table.columns, row) { + let text_raw = line_to_html!(cell).trim().to_string(); + let text_cmp = text_raw.to_lowercase().to_string(); + let text = match text_cmp.as_str() { + "yes" => "✓", + "no" => "✗", + _ => &text_raw, + }; + let mut class = match text { + "--" => "c", + _ => match column.alignment { + Alignment::Left => "l", + Alignment::Center => "c", + Alignment::Right => "r", + }, + }.to_string(); + if ["no", "--", "0"].contains(&text_cmp.as_str()) { + class.push_str(" dim"); + }; + if column.border_right { + class.push_str(" border"); + } + html!("<td class='{class}'>{text}</td>"); + }) + }) + }; + })) + } } - } + ); return html; } -fn line_to_html(line: &Line, page: &Page, website: &Website) -> String { +fn line_to_html(line: &Line, page: &Page, website: &Website, prefix: &Option<String>) -> String { let mut html = String::new(); for line_element in &line.tokens { match line_element { @@ -333,7 +404,7 @@ fn line_to_html(line: &Line, page: &Page, website: &Website) -> String { Token::Math(text) => { let text = &sanitize_text(text, false); html.push_str(&format!("<span class='math'>{text}</span>")) } Token::InternalLink{ label: link_label, path } => { - let ParsedLink { path, class, mut label } = parse_internal_link(path, page, website); + let ParsedLink { path, class, mut label } = parse_internal_link(path, page, website, prefix); if !link_label.is_empty() { label = link_label.to_string(); } @@ -356,30 +427,56 @@ struct ParsedLink { pub class: &'static str, } -fn parse_internal_link(name: &str, page: &Page, website: &Website) -> ParsedLink { +fn parse_internal_link(name: &str, page: &Page, website: &Website, prefix: &Option<String>) -> ParsedLink { let from = &page; let (class, label, path) = match name.split_once('#') { - Some(("", heading)) => ("heading", heading, format!("#{}", strip_appendix(heading))), - Some((page, heading)) => ("page", heading, format!("{page}.html#{}", strip_appendix(heading))), + Some(("", heading)) => ("heading", heading, format!("#{heading}")), + Some((page, heading)) => ("page", heading, format!("{page}.html#{heading}")), _ => ("page", name, format!("{name}.html")), }; - let mut path = make_url_safe(&path); + let mut path = to_slug(&path); let label = match label.rsplit_once('/') { Some((_, label)) => sanitize_text(label.trim(), true), None => sanitize_text(label.trim(), true), }; - // Check that the linked internal page exists. + // Check that the linked internal page with heading exists. if class == "page" { match website.has_page(page, &path, "html") { Some(resolved) => path = resolved, None => warn!("Page {from:?} contains link to nonexistent page {path:?}"), } } - // Check that the heading exists. + // Check that the heading exists on this page. if class == "heading" { - let heading = path.strip_prefix('#').unwrap(); - if !page.headings.iter().any(|h| h.url == heading) { - warn!("Page {from:?} contains link to nonexistent internal heading {heading:?}"); + let plain_heading = path.strip_prefix('#').unwrap().to_string(); + let prefixed_heading = match prefix { + Some(prefix) => format!("{prefix}-{plain_heading}"), + None => plain_heading.to_string(), + }; + let mut matched = false; + for heading in &page.headings { + if heading.name.slug() == plain_heading { + if heading.prefix.is_some() { + // The matched heading has a prefix, so is one of many. + // The prefix must match, we must disambiguate the path. + if heading.slug() == prefixed_heading { + matched = true; + path = format!("#{prefixed_heading}"); + break; + } + } else { + // The matched heading has no prefix, so is unique on the page. + matched = true; + break + } + } + } + if !matched { + let prefix_note = match prefix { + Some(prefix) => format!(" (under {prefix:?})"), + None => format!(""), + }; + warn!("Page {from:?} contains link to nonexistent internal heading {plain_heading:?}{prefix_note}"); } } let path = url_encode(&path); @@ -418,63 +515,3 @@ fn parse_external_link(label: &str, path: &str, page: &Page, website: &Website) let label = sanitize_text(&label, true); ParsedLink { path, class: "external", label } } - - -/// Replace each HTML-reserved character with an HTML-escaped character. -fn sanitize_text(text: &str, fancy: bool) -> String { - let mut output = String::new(); - let chars: Vec<char> = text.chars().collect(); - for (i, c) in chars.iter().enumerate() { - let prev = match i > 0 { - true => chars[i - 1], - false => ' ', - }; - let next = match i + 1 < chars.len() { - true => chars[i + 1], - false => ' ', - }; - match c { - '&' => { - // The HTML syntax for unicode characters is � - if let Some('#') = chars.get(i+1) { output.push(*c) } - else { output.push_str("&") } - }, - '<' => output.push_str("<"), - '>' => output.push_str(">"), - '"' => match fancy { - true => match prev.is_whitespace() { - true => output.push('“'), - false => output.push('”'), - } - false => output.push_str("""), - }, - '\'' => match fancy { - true => match prev.is_whitespace() { - true => output.push('‘'), - false => output.push('’'), - } - false => output.push_str("'"), - }, - '-' if fancy => match prev.is_whitespace() && next.is_whitespace() { - true => match i > 0 { - true => output.push('—'), // em-dash, for mid-sentence - false => output.push('–'), // en-dash, for start of line - } - false => output.push('-'), // regular dash, for mid-word - } - _ => output.push(*c), - } - } - return output; -} - - -/// Remove a 'Appendix #: ' prefix from a string. -pub fn strip_appendix(text: &str) -> &str { - if let Some((prefix, name)) = text.split_once(": ") { - if prefix.starts_with("Appendix") { - return name; - } - } - return text; -} diff --git a/src/generate_rss.rs b/src/generate_rss.rs new file mode 100644 index 0000000..48a6917 --- /dev/null +++ b/src/generate_rss.rs @@ -0,0 +1,72 @@ +use crate::*; + +use std::collections::VecDeque; + +use chrono::{DateTime, Utc, Local}; + + +pub fn generate_rss(feed: &Feed, website: &Website) -> String { + let path = &feed.source_path; + let content = std::fs::read_to_string(path).unwrap(); + let mut lines: VecDeque<&str> = content.lines().collect(); + let base_url = website.config.get("rss.base_url"); + if base_url.is_empty() { + warn!("No value was given for 'rss.base_url' key in toaster.conf"); + } + let (parent_url, _) = feed.url.split_once('/').unwrap(); + + let channel_title = lines.pop_front().unwrap_or("No title"); + let last_build_date = match feed.last_modified { + Some(system_time) => system_time.into(), + None => Utc::now(), + }.to_rfc2822(); + let mut all_entries = String::new(); + + for line in &lines { + if line.is_empty() { continue } + if let Some((timestamp, name)) = line.split_once("::") { + let mut timestamp = timestamp.to_string(); + let entry_title = name; + if !timestamp.contains('T') { + timestamp.push_str("T00:00:00"); + } + if !timestamp.contains('Z') || timestamp.contains('+') { + let offset = Local::now().offset().to_string(); + timestamp.push_str(&offset); + } + let Ok(entry_time) = DateTime::parse_from_rfc3339(×tamp) else { + warn!("Invalid timestamp in RSS file {path:?}: {timestamp:?}"); + continue; + }; + let entry_link = format!("{base_url}/{parent_url}/{}.html", to_slug(name)); + + // Check that child page exists. + if let None = website.has_page(feed, name, "html") { + warn!("Feed {feed:?} contains link to nonexistent page {name:?}"); + } + + + let entry_string = format!(" + <item> + <title>{entry_title}</title> + <link>{entry_link}</link> + <pubDate>{entry_time}</pubDate> + </item> +"); + all_entries.push_str(&entry_string); + } else { + warn!("Invalid line in RSS file {path:?}: {line:?}"); + } + } + + format!( +r#"<?xml version="1.0" encoding="UTF-8" ?> +<rss version="2.0"> + <channel> + <title>{channel_title}</title> + <link>{base_url}</link> + <lastBuildDate>{last_build_date}</lastBuildDate> +{all_entries} + </channel> +</rss>"#) +} diff --git a/src/main.rs b/src/main.rs index 1ea25d2..dad8377 100644 --- a/src/main.rs +++ b/src/main.rs @@ -1,9 +1,16 @@ #![feature(path_add_extension)] mod collect_files; -pub use collect_files::*; +mod config; mod generate_html; +mod generate_rss; +mod string_utils; + +pub use collect_files::*; +pub use config::*; pub use generate_html::*; +pub use generate_rss::*; +pub use string_utils::*; use markdown::*; use vagabond::*; @@ -12,81 +19,97 @@ use std::collections::HashSet; use std::time::SystemTime; use log::{info, warn, error, fatal}; -use switchboard::{Switchboard, SwitchQuery}; - - -fn print_help() -> ! { - eprintln!("\ -Usage: toaster <source> <destination> - -Generate a website from a structured directory of markdown files. - -Arguments: - source Source directory with markdown files - destination Path to output directory - -Switches: - --delete Delete the destination directory first if it exists - --html Generate HTML output - --version, -v Print information as each file is parsed - --version Print the program version and exit - --help, -h Print help -"); - std::process::exit(0); -} - -fn print_version() -> ! { - let version = env!("CARGO_PKG_VERSION"); - eprintln!("toaster, version {version}"); - eprintln!("written by ben bridle"); - std::process::exit(0); -} +use switchboard::*; fn main() { let mut args = Switchboard::from_env(); - if args.named("help").short('h').as_bool() { + + // Informational switches. + args.named("help").short('h'); + args.named("version"); + args.named("verbose").short('v'); + if args.get("help").as_bool() { print_help(); + std::process::exit(0); } - if args.named("version").as_bool() { - print_version(); + if args.get("version").as_bool() { + let version = env!("CARGO_PKG_VERSION"); + eprintln!("toaster, version {version}"); + eprintln!("written by ben bridle"); + std::process::exit(0); } - if args.named("verbose").short('v').as_bool() { + if args.get("verbose").as_bool() { log::set_log_level(log::LogLevel::Info); } - let source = args.positional("source").as_path(); - let destination = args.positional("destination").as_path(); - let delete_existing = args.named("delete").as_bool(); - let export_html = args.named("html").as_bool(); + // Functional switches. + args.positional("source"); + args.positional("destination"); + args.named("delete"); + args.named("html"); + args.named("use-symlinks"); + args.raise_errors(); + let source = args.get("source").as_path(); + let destination = args.get("destination").as_path(); + let delete_existing = args.get("delete").as_bool(); + let export_html = args.get("html").as_bool(); + let use_symlinks = args.get("use-symlinks").as_bool(); + #[cfg(not(target_family = "unix"))] + if use_symlinks { + fatal!("Symbolic links are only supported on Linux"); + } + + // Parse entire website directory. let source = match source.canonicalize() { Ok(source) => source, Err(err) => fatal!("{source:?}: {err}"), }; - let website = Website::from_path(&source); + // ------------------------------------------------------------ + // Check for duplicate output paths for pages. - let mut destinations: HashSet<&str> = HashSet::new(); + let mut urls: HashSet<&str> = HashSet::new(); let mut duplicates: HashSet<&str> = HashSet::new(); for page in &website.pages { - if !destinations.insert(&page.full_url) { - duplicates.insert(&page.full_url); + if !urls.insert(&page.url) { + duplicates.insert(&page.url); + }; + } + for static_file in &website.static_files { + if !urls.insert(&static_file.url) { + duplicates.insert(&static_file.url); + }; + } + for redirect in &website.redirects { + if !urls.insert(&redirect.url) { + duplicates.insert(&redirect.url); }; } if !duplicates.is_empty() { for destination in duplicates { - warn!("Multiple pages have the output path {destination:?}"); + warn!("Multiple files, pages, or redirects have the output path {destination:?}"); for page in &website.pages { - if page.full_url == destination { + if page.url == destination { eprintln!(":: {:?}", page.source_path); } } + for static_file in &website.static_files { + if static_file.url == destination { + eprintln!(":: {:?}", static_file.source_path); + } + } + for redirect in &website.redirects { + if redirect.url == destination { + eprintln!(":: {:?}", redirect.source_path); + } + } } } let mut destination = destination; - destination.push(make_url_safe(&website.name)); + destination.push(&website.name.slug()); if delete_existing && Entry::from_path(&destination).is_ok() { info!("Deleting existing destination directory {destination:?}"); @@ -97,10 +120,10 @@ fn main() { for page in &website.pages { let mut destination = destination.clone(); - destination.push(&page.full_url); + destination.push(&page.url); // Convert document to different formats. if export_html { - let html = generate_html(&page.document, page, &website); + let html = generate_html(page, &website); write_file(&html, &destination, "html", page.last_modified); } // Copy original markdown file. @@ -113,35 +136,70 @@ fn main() { for static_file in &website.static_files { let mut destination = destination.clone(); - destination.push(&static_file.full_url); - info!("Copying static file to {destination:?}"); - make_parent_directory(&destination).unwrap(); - copy(&static_file.source_path, &destination).unwrap_or_else(|_| - error!("Failed to copy static file {:?} to {:?}", - static_file.source_path, destination)); + destination.push(&static_file.url); + if use_symlinks { + #[cfg(target_family = "unix")] + { + info!("Linking static file to {destination:?}"); + make_parent_directory(&destination).unwrap(); + let _ = remove(&destination); + std::os::unix::fs::symlink(&static_file.source_path, &destination).unwrap_or_else(|_| + error!("Failed to link static file {:?} to {:?}", + static_file.source_path, destination)); + } + } else { + info!("Copying static file to {destination:?}"); + copy(&static_file.source_path, &destination).unwrap_or_else(|_| + error!("Failed to copy static file {:?} to {:?}", + static_file.source_path, destination)); + } + } // NOTE: Static dir contents are copied as part of all static files. for redirect in &website.redirects { let mut destination = destination.clone(); - destination.push(&redirect.full_url); - let path = &redirect.redirect; + destination.push(&redirect.url); + let path = &redirect.target; if export_html { if !path.contains("://") { if let Some(path) = website.has_page(redirect, &path, "html") { - write_file(&generate_html_redirect(&path), &destination, "html", redirect.last_modified); + write_file(&generate_html_redirect(&path, &website), &destination, "html", redirect.last_modified); } else { warn!("Redirect {:?} links to nonexistent page {path:?}", redirect.name); } } else { - write_file(&generate_html_redirect(&path), &destination, "html", redirect.last_modified); + write_file(&generate_html_redirect(&path, &website), &destination, "html", redirect.last_modified); } } } + + for feed in &website.feeds { + let mut destination = destination.clone(); + destination.push(&feed.url); + write_file(&generate_rss(feed, &website), &destination, "rss", feed.last_modified); + } } +fn print_help() { + eprintln!("\ +Usage: toaster <source> <destination> +Generate a website from a structured directory of markdown files. + +Arguments: + source Source directory with markdown files + destination Path to output directory + +Switches: + --delete Delete the destination directory first if it exists + --html Generate HTML output + --version, -v Print information as each file is parsed + --version Print the program version and exit + --help, -h Print help +"); +} pub fn write_file(text: &str, destination: &PathBuf, ext: &str, last_modified: Option<SystemTime>) { let mut destination = destination.clone(); @@ -158,23 +216,3 @@ pub fn write_file(text: &str, destination: &PathBuf, ext: &str, last_modified: O } } } - -pub fn make_url_safe(text: &str) -> String { - text.to_ascii_lowercase().chars().filter_map(|c| - if c.is_alphanumeric() || "-_~.+/#".contains(c) { Some(c) } - else if c == ' ' { Some('-') } - else { None } ) - .collect() -} - -pub fn url_encode(text: &str) -> String { - let mut output = String::new(); - for c in text.chars() { - match c { - '"' => output.push_str("%22"), - '\'' => output.push_str("%27"), - _ => output.push(c), - } - } - return output; -} diff --git a/src/string_utils.rs b/src/string_utils.rs new file mode 100644 index 0000000..cceb575 --- /dev/null +++ b/src/string_utils.rs @@ -0,0 +1,145 @@ +use crate::*; + + +#[derive(Clone)] +pub struct Name { + raw: String, +} + +impl Name { + /// Preserve markdown syntax, return raw string. + pub fn raw(&self) -> String { + self.raw.clone() + } + /// Parse markdown syntax, return styled line. + pub fn styled(&self) -> Line { + Line::from_str(&self.raw) + } + /// Strip out markdown syntax, return plain text. + pub fn plain(&self) -> String { + self.styled().to_string() + } + /// Strip out markdown syntax, return url-safe text. + pub fn slug(&self) -> String { + to_slug(&self.plain()) + } +} + +impl std::fmt::Display for Name { + fn fmt(&self, f: &mut std::fmt::Formatter) -> Result<(), std::fmt::Error> { + self.raw.fmt(f) + } +} + +impl std::fmt::Debug for Name { + fn fmt(&self, f: &mut std::fmt::Formatter) -> Result<(), std::fmt::Error> { + self.raw.fmt(f) + } +} + +impl PartialEq for Name { + fn eq(&self, other: &string_utils::Name) -> bool { + self.slug() == other.slug() + } +} + +impl Eq for Name {} +impl std::hash::Hash for Name { + fn hash<H>(&self, hasher: &mut H) where H: std::hash::Hasher { + self.slug().hash(hasher) + } +} + +impl From<String> for Name { + fn from(raw: String) -> Self { + Self { raw } + } +} + +impl From<&str> for Name { + fn from(raw: &str) -> Self { + Self { raw: raw.to_string() } + } +} + + + +// Turn a string into a tidy URL slug. +pub fn to_slug(text: &str) -> String { + let mut string = String::new(); + let mut prev = ' '; + for c in text.to_lowercase().chars() { + let c = match c == ' ' { + true => '-', + false => c, + }; + if c.is_alphanumeric() { string.push(c) } + if "_~.+/#".contains(c) { string.push(c) } + if c == '-' && prev != '-' { string.push(c) } + prev = c; + } + return string; +} + +// Prevent link hrefs from breaking out of quotations. +pub fn url_encode(text: &str) -> String { + let mut output = String::new(); + for c in text.chars() { + match c { + '"' => output.push_str("%22"), + '\'' => output.push_str("%27"), + _ => output.push(c), + } + } + return output; +} + +/// Replace each HTML-reserved character with an HTML-escaped character. +pub fn sanitize_text(text: &str, fancy: bool) -> String { + let mut output = String::new(); + let chars: Vec<char> = text.chars().collect(); + for (i, c) in chars.iter().enumerate() { + let prev = match i > 0 { + true => chars[i - 1], + false => ' ', + }; + let next = match i + 1 < chars.len() { + true => chars[i + 1], + false => ' ', + }; + let is_whitespace = |c: char| c.is_whitespace() || "()[].,".contains(c); + + match c { + '&' => { + // The HTML syntax for unicode characters is � + if let Some('#') = chars.get(i+1) { output.push(*c) } + else { output.push_str("&") } + }, + '<' => output.push_str("<"), + '>' => output.push_str(">"), + '"' => match fancy { + true => match is_whitespace(prev) { + true => output.push('“'), + false => output.push('”'), + } + false => output.push_str("""), + }, + '\'' => match fancy { + true => match is_whitespace(prev) { + true => output.push('‘'), + false => output.push('’'), + } + false => output.push_str("'"), + }, + '-' if fancy => match prev.is_whitespace() && next.is_whitespace() { + true => match i > 0 { + true => output.push('—'), // em-dash, for mid-sentence + false => output.push('–'), // en-dash, for start of line + } + false => output.push('-'), // regular dash, for mid-word + } + _ => output.push(*c), + } + } + return output; +} |