support for additional linters (link-checker & aspell) (#410)

* Added support for additional linters (html-proofer link checker) - added additional test matrix item just for linters - install and build mdbook only when needed to speedup CI - reorganized the travis scripts * Added spellchecking script from rust-book to CI also fixed minor typos * updated and moved serde_json links to propper position * move link checking to link-checker
2024-11-24 20:43:07 +00:00 · 2018-06-06 17:33:25 +02:00 · 2018-06-06 17:33:25 +02:00 · 0c167cb32c
commit 0c167cb32c
parent af25e615d2
11 changed files with 554 additions and 24 deletions
--- a/.travis.yml
+++ b/.travis.yml
@ -12,8 +12,29 @@ env:
  global:
  - secure: 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

+matrix:
+  include:
+    - rust: stable
+      os: linux
+      env: CONTENT_TESTS=1
+    - rust: stable
+      os: linux
+      env: CONTENT_TESTS=1 CONTENT_TESTS_LINKS=1
+  allow_failures:
+    - rust: stable
+      os: linux
+      env: CONTENT_TESTS=1 CONTENT_TESTS_LINKS=1
+
+addons:
+  apt:
+    packages:
+      - aspell
+      - aspell-en
+
 before_install:
-  - cargo install mdbook --vers '0.1.7' --debug
+  - ./ci/install_deps.sh
  - export PATH=$HOME/.cargo/bin:$PATH

-after_success: ./deploy.sh
+script: ./ci/test_script.sh
+
+after_success: ./ci/deploy.sh
--- a/ci/deploy.sh
+++ b/ci/deploy.sh
@ -2,6 +2,8 @@

 set -o errexit -o nounset

+echo "Running $0"
+
 if [ -z "${TRAVIS_BRANCH:-}" ]; then
    echo "This script may only be run from Travis!"
    exit 1
@ -14,16 +16,22 @@ if [[ "$TRAVIS_BRANCH" != "master" || "$TRAVIS_RUST_VERSION" != "stable" || "$TR
    exit 0
 fi

-# check for outdated dependencies on nightly builds
-if [ "${TRAVIS_EVENT_TYPE:-}" == "cron" ]; then
-    echo "This is cron build. Checking for outdated dependencies!"
-    rm ./Cargo.lock
-    cargo clean
-    # replace all [dependencies] versions with "*"
-    sed -i -e "/^\[dependencies\]/,/^\[.*\]/ s|^\(.*=[ \t]*\).*$|\1\"\*\"|" ./Cargo.toml

-    cargo test || { echo "Cron build failed! Dependencies outdated!"; exit 1; }
-    echo "Cron build success! Dependencies are up to date!"
+if [ -z "${CONTENT_TESTS:-}" ]; then
+    # check for outdated dependencies on nightly builds
+    if [ "${TRAVIS_EVENT_TYPE:-}" == "cron" ]; then
+        echo "This is cron build. Checking for outdated dependencies!"
+        rm ./Cargo.lock
+        cargo clean
+        # replace all [dependencies] versions with "*"
+        sed -i -e "/^\[dependencies\]/,/^\[.*\]/ s|^\(.*=[ \t]*\).*$|\1\"\*\"|" ./Cargo.toml
+
+        cargo test || { echo "Cron build failed! Dependencies outdated!"; exit 1; }
+        echo "Cron build success! Dependencies are up to date!"
+        exit 0
+    fi
+
+    echo "We deploy only after we also test the markup and descriptions!"
    exit 0
 fi

--- a/ci/dictionary.txt
+++ b/ci/dictionary.txt
@ -0,0 +1,328 @@
+personal_ws-1.1 en 0 utf-8
+abcde
+ABCDEFGHIJKLMNOPQRSTUVWXYZ
+abcdefghijklmnopqrstuvwxyz
+Addr
+addr
+AddrParse
+AddrParseError
+AfterPath
+Akshat
+akshat
+alisha
+AlphaNumeric
+ApiResponse
+Appender
+appender
+args
+ascii
+ashley
+AsRef
+attr
+Auth
+auth
+backend
+BACKTRACE
+Backtrace
+backtrace
+BeforePath
+bitfield
+bitflags
+bitwise
+bool
+BufRead
+BufReader
+byteorder
+ByteRecord
+bytestring
+cAFeEDEcafBAd
+CannotBeABase
+ccccff
+ChainedError
+CHARSET
+chrono
+ClientBuilder
+CmdError
+colombo
+concat
+Config
+config
+ConsoleLogger
+const
+ContentRange
+ContentRangeSpec
+ContentType
+Cookin
+cpus
+CString
+CSVError
+CsvError
+CsvInnerError
+ctrl
+Datelike
+DateParse
+DateTime
+datetime
+DEadBEEfc
+DecodeError
+dedup
+deduplicated
+delhi
+Deref
+Deserialize
+deserialize
+deserialized
+Deserializer
+deserializer
+dest
+DirEntry
+dotfiles
+DynamicImage
+endian
+endif
+EnvLogger
+EnvVar
+eprintln
+ErrorKind
+extern
+ferris
+FileAppender
+filename
+filenames
+filepath
+filesystem
+FilterType
+FixedOffset
+flate
+fname
+Formatter
+foundational
+fout
+FromStr
+FromStrError
+FromUtf
+gists
+GitHub
+github
+GlobError
+Guybrush
+GzDecoder
+GzEncoder
+Hackerman
+Hardcoded
+hardcoded
+HashMap
+HashSet
+hashtag
+HASHTAG
+hashtags
+headders
+HeadersEcho
+hexa
+HexColor
+HEXUPPER
+HMAC
+hmac
+Hong
+href
+html
+http
+HttpRequest
+ifdef
+ImageBuffer
+ImageError
+impl
+incrementing
+IndependentSample
+init
+initialised
+IntoInnerError
+IntoIter
+IntoIterator
+io's
+IOError
+IoError
+IpAddr
+ITER
+iter
+Iterable
+iterable
+Janeiro
+journalctl
+JSON
+Json
+json
+julia
+lang
+LevelFilter
+libhello
+Libz
+linux
+LittleEndian
+LogBuilder
+LogConfig
+logfile
+login
+LogLevel
+LogLevelFilter
+LogMetadata
+LogRecord
+LogTarget
+loopback
+MatchOptions
+MediaWiki
+mediawiki
+memmap
+Metadata
+metadata
+mkdir
+Mmap
+mpsc
+MpscRecv
+Mutex
+MutexGuard
+myfile
+MyFlags
+NaiveDate
+NaiveDateTime
+NaiveTime
+nFun
+NotFound
+NulError
+OAuth
+oneline
+ParallelIterator
+parallelized
+params
+ParseError
+ParseInt
+ParseIntError
+ParseOptions
+PartialContent
+PartialEq
+PartialRangeIter
+passwd
+PathBuf
+PatternEncoder
+PatternError
+PBKDF
+pbkdf
+PhoneNumber
+PNGs
+preallocating
+prerelease
+prev
+printf
+println
+proc
+programmatically
+radix
+RandomResponseError
+RateLimit
+ReadBytesExt
+ReaderBuilder
+recusively
+recv
+RecvError
+REGEX
+Regex
+regex
+RegexSet
+RegexSetBuilder
+Replacer
+repo
+reponse
+ReqError
+ReqParseError
+RequestBuilder
+Reqwest
+reqwest
+resize
+resized
+RESTful
+ReverseDependencies
+rustaceans
+rustc
+RwLock
+SecureRandom
+SemVer
+Semver
+semver
+SemVerError
+SemVerReq
+Serde
+serde
+SetLogger
+SetLoggerError
+SigningKey
+SocketAddrV
+StatusCode
+stderr
+stdin
+Stdout
+stdout
+strftime
+StringRecord
+StripPrefixError
+strs
+struct
+structs
+subcommands
+subdirectories
+subfolders
+subprocess
+subtype
+sydney
+symlinks
+syslog
+SyslogError
+SystemRandom
+SystemTime
+SystemTimeError
+TcpListener
+TcpStream
+TempDir
+tempdir
+ThreadPool
+threadpool
+Threepwood
+Timelike
+Timestamp
+timestamp
+timezones
+TOML
+Toml
+toml
+tuple
+Tuple
+typesafe
+unary
+unwinded
+UpperHex
+uptime
+urlencoded
+UrlParse
+UrlParseError
+urls
+UserAgent
+userid
+usize
+VarError
+versa
+Versioning
+VersionReq
+vreq
+WalkDir
+Walkdir
+walkdir
+webpage
+webservice
+whitespace
+WriteBytesExt
+XPoweredBy
+XRateLimitLimit
+XRateLimitRemaining
+XRateLimitReset
+YAML
+YYYY
+zurich
--- a/ci/install_deps.sh
+++ b/ci/install_deps.sh
@ -0,0 +1,23 @@
+#!/bin/bash
+
+set -o errexit -o nounset
+
+echo "Running $0"
+
+if [ -z "${TRAVIS_BRANCH:-}" ]; then
+    echo "This script may only be run from Travis!"
+    exit 1
+fi
+
+if [[ "${CONTENT_TESTS:-}" == 1 ]]; then
+    echo "Installing additional dependencies"
+
+    if [[ "${CONTENT_TESTS_LINKS:-}" == 1 ]]; then
+        pyenv global system 3.6
+        pip3 install --user link-checker==0.1.0
+    fi
+    cargo install mdbook --vers '0.1.7' --debug
+fi
+
+exit 0
+
--- a/ci/spellcheck.sh
+++ b/ci/spellcheck.sh
@ -0,0 +1,102 @@
+#!/bin/bash
+# Copyright 2016 The Rust Project Developers. See the COPYRIGHT
+# file at the top-level directory of this distribution and at
+# http://rust-lang.org/COPYRIGHT.
+#
+# Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
+# http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
+# <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
+# option. This file may not be copied, modified, or distributed
+# except according to those terms.
+
+aspell --version
+
+# Checks project markdown files for spell errors
+
+# Notes:
+
+# This script needs dictionary file ($dict_filename) with project-specific
+# valid words. If this file is missing, first invocation of a script generates
+# a file of words considered typos at the moment. User should remove real typos
+# from this file and leave only valid words. When script generates false
+# positive after source modification, new valid word should be added
+# to dictionary file.
+
+# Default mode of this script is interactive. Each source file is scanned for
+# typos. aspell opens window, suggesting fixes for each found typo. Original
+# files with errors will be backed up to files with format "filename.md.bak".
+
+# When running in CI, this script should be run in "list" mode (pass "list"
+# as first argument). In this mode script scans all files and reports found
+# errors. Exit code in this case depends on scan result:
+# 1 if any errors found,
+# 0 if all is clear.
+
+# Script skips words with length less than or equal to 3. This helps to avoid
+# some false positives.
+
+# We can consider skipping source code in markdown files (```code```) to reduce
+# rate of false positives, but then we lose ability to detect typos in code
+# comments/strings etc.
+
+shopt -s nullglob
+
+dict_filename=./ci/dictionary.txt
+markdown_sources=(./src/*.md)
+mode="check"
+
+# aspell repeatedly modifies personal dictionary for some purpose,
+# so we should use a copy of our dictionary
+mkdir -p "/tmp/ci"
+dict_path="/tmp/$dict_filename"
+
+if [[ "$1" == "list" ]]; then
+    mode="list"
+fi
+
+if [[ ! -f "$dict_filename" ]]; then
+    # Pre-check mode: generates dictionary of words aspell consider typos.
+    # After user validates that this file contains only valid words, we can
+    # look for typos using this dictionary and some default aspell dictionary.
+    echo "Scanning files to generate dictionary file '$dict_filename'."
+    echo "Please check that it doesn't contain any misspellings."
+
+    echo "personal_ws-1.1 en 0 utf-8" > "$dict_filename"
+    cat "${markdown_sources[@]}" | aspell --ignore 3 list | sort -u >> "$dict_filename"
+elif [[ "$mode" == "list" ]]; then
+    # List (default) mode: scan all files, report errors
+    declare -i retval=0
+
+    cp "$dict_filename" "$dict_path"
+
+    if [ ! -f $dict_path ]; then
+        retval=1
+        exit "$retval"
+    fi
+
+    for fname in "${markdown_sources[@]}"; do
+        command=$(aspell --ignore 3 --personal="$dict_path" "$mode" < "$fname")
+        if [[ -n "$command" ]]; then
+            for error in $command; do
+                # FIXME: Find more correct way to get line number
+                # (ideally from aspell). Now it can make some false positives,
+                # because it is just a grep
+                grep --with-filename --line-number --color=always "$error" "$fname"
+            done
+            retval=1
+        fi
+    done
+    exit "$retval"
+elif [[ "$mode" == "check" ]]; then
+    # Interactive mode: fix typos
+    cp "$dict_filename" "$dict_path"
+
+    if [ ! -f $dict_path ]; then
+        retval=1
+        exit "$retval"
+    fi
+
+    for fname in "${markdown_sources[@]}"; do
+        aspell --ignore 3 --dont-backup --personal="$dict_path" "$mode" "$fname"
+    done
+fi
--- a/ci/test_script.sh
+++ b/ci/test_script.sh
@ -0,0 +1,51 @@
+#!/bin/bash
+
+set -o errexit -o nounset
+
+echo "Running $0"
+
+if [ -z "${TRAVIS_BRANCH:-}" ]; then
+    echo "This script may only be run from Travis!"
+    exit 1
+fi
+
+# Returns 1 if program is installed and 0 otherwise
+program_installed() {
+    local return_=1
+
+    type $1 >/dev/null 2>&1 || { local return_=0; }
+
+    echo "$return_"
+}
+
+if [[ "${CONTENT_TESTS:-}" == 1 ]]; then
+    # Ensure required programs are installed
+    if [ $(program_installed mdbook) == 0 ]; then
+        echo "Please install mdbook: cargo install mdbook."
+        exit 1
+    fi
+
+    echo "Testing markup and descriptions"
+    echo "Building site to book/"
+    mdbook build
+    echo "Checking spelling"
+    ./ci/spellcheck.sh list
+
+    if [[ "${CONTENT_TESTS_LINKS:-}" == 1 ]]; then
+        if [ $(program_installed link-checker) == 0 ]; then
+            echo "Please install link-checker: 'pip install link-checker'"
+            exit 1
+        fi
+        echo "Checking local links:"
+        # failing local link test  is a hard error as there should be no false possitives
+        link-checker --no-external ./book/
+
+        echo "Checking external links:"
+        # we do not want to fail on false positives
+        link-checker --no-local ./book/ --ignore '.*api.github.com*.' || true
+    fi
+else
+    echo "Testing code snippets"
+    cargo build --verbose
+    cargo test --verbose
+fi
--- a/src/app.md
+++ b/src/app.md
@ -151,7 +151,7 @@ fn run() -> Result<()> {
    let tar = GzDecoder::new(tar_gz);
    // Load the archive from the tarball
    let mut archive = Archive::new(tar);
-    // Unpack the archive inside curent working directory
+    // Unpack the archive inside current working directory
    archive.unpack(".")?;

    Ok(())
--- a/src/basics.md
+++ b/src/basics.md
@ -1585,7 +1585,7 @@ Parses a [`DateTime`] struct from strings representing the well-known formats

 Escape sequences that are available for the [`DateTime::parse_from_str`] can be
 found at [`chrono::format::strftime`]. Note that the [`DateTime::parse_from_str`]
-requires that such a DateTime struct must be creatable that it uniquely
+requires that such a DateTime struct can be created that it uniquely
 identifies a date and a time. For parsing dates and times without timezones use
 [`NaiveDate`], [`NaiveTime`], and [`NaiveDateTime`].

@ -1720,7 +1720,7 @@ fn main() {

 Gets the current working directory by calling [`env::current_dir`],
 then for each entries in [`fs::read_dir`], extracts the
-[`DirEntry::path`] and gets the metada via [`fs::Metadata`]. The
+[`DirEntry::path`] and gets the metadata via [`fs::Metadata`]. The
 [`Metadata::modified`] returns the [`SystemTime::elapsed`] time since
 last modification of the entry. It's converted into seconds with
 [`Duration::as_secs`] and compared with 24 hours (24 * 60 * 60
--- a/src/build_tools.md
+++ b/src/build_tools.md
@ -179,7 +179,7 @@ fn main(){

 It is simple to build bundled C code with custom defines using [`cc::Build::define`]. 
 It takes an [`Option`] value, so it is possible to create defines such as `#define APP_NAME "foo"`
-as well as `#define WELCOME` (pass `None` as the value for a value-less defne). This example builds
+as well as `#define WELCOME` (pass `None` as the value for a value-less define). This example builds
 a bundled C file with dynamic defines set in `build.rs` and prints "**Welcome to foo - version 1.0.2**"
 when run. Cargo sets some [environment variables][cargo-env] which may be useful for some custom defines.

--- a/src/encoding.md
+++ b/src/encoding.md
@ -23,22 +23,16 @@

 [![serde-json-badge]][serde-json] [![cat-encoding-badge]][cat-encoding]

-The [`serde_json`] crate provides a [`from_str`] function to parse a `&str` of
+The [serde_json] crate provides a [`serde_json::from_str`] function to parse a `&str` of
 JSON into a type of the caller's choice.

-[`serde_json`]: https://docs.serde.rs/serde_json/
-[`from_str`]: https://docs.serde.rs/serde_json/fn.from_str.html
-
 Unstructured JSON can be parsed into a universal [`serde_json::Value`] type that
 is able to represent any valid JSON data.

-[`serde_json::Value`]: https://docs.serde.rs/serde_json/enum.Value.html
-
 The example below shows a `&str` of JSON being parsed and then compared to what
 we expect the parsed value to be. The expected value is declared using the
 [`json!`] macro.

-[`json!`]: https://docs.serde.rs/serde_json/macro.json.html

 ```rust
 # #[macro_use]
@ -831,18 +825,21 @@ fn main() {
 <!-- API Reference -->

 [`csv::ByteRecord`]: https://docs.rs/csv/*/csv/struct.ByteRecord.html
+[`csv::invalid_option`]: https://docs.rs/csv/*/csv/fn.invalid_option.html
 [`csv::Reader::deserialize`]: https://docs.rs/csv/*/csv/struct.Reader.html#method.deserialize
 [`csv::Reader::deserialize`]: https://docs.rs/csv/\*/csv/struct.Reader.html#method.deserialize
 [`csv::StringRecord`]: https://docs.rs/csv/*/csv/struct.StringRecord.html
 [`csv::Writer`]: https://docs.rs/csv/*/csv/struct.Writer.html
-[`csv::invalid_option`]: https://docs.rs/csv/*/csv/fn.invalid_option.html
 [`flush`]: https://docs.rs/csv/*/csv/struct.Writer.html#method.flush
 [`form_urlencoded::byte_serialize`]: https://docs.rs/url/*/url/form_urlencoded/fn.byte_serialize.html
 [`form_urlencoded::parse`]: https://docs.rs/url/*/url/form_urlencoded/fn.parse.html
 [`FromStrError`]: https://docs.rs/mime/*/mime/struct.FromStrError.html
+[`json!`]: https://docs.rs/serde_json/*/serde_json/macro.json.html
 [`MIME`]: https://docs.rs/mime/*/mime/struct.Mime.html
 [`percent_decode`]: https://docs.rs/percent-encoding/*/percent_encoding/fn.percent_decode.html
 [`serde::Deserialize`]: https://docs.rs/serde/\*/serde/trait.Deserialize.html
+[`serde_json::from_str`]: https://docs.rs/serde_json/*/serde_json/fn.from_str.html
+[`serde_json::Value`]: https://docs.rs/serde_json/*/serde_json/enum.Value.html
 [`serialize`]: https://docs.rs/csv/*/csv/struct.Writer.html#method.serialize
 [`std::str::FromStr`]: https://doc.rust-lang.org/std/str/trait.FromStr.html
 [`utf8_percent_encode`]: https://docs.rs/percent-encoding/*/percent_encoding/fn.utf8_percent_encode.html
--- a/src/links.md
+++ b/src/links.md
@ -110,7 +110,7 @@ Keep lines sorted.
 [semver]: https://docs.rs/semver/
 [serde-badge]: https://badge-cache.kominick.com/crates/v/serde.svg?label=serde
 [serde-json-badge]: https://badge-cache.kominick.com/crates/v/serde_json.svg?label=serde_json
-[serde-json]: https://docs.serde.rs/serde_json/
+[serde-json]: https://docs.rs/serde_json/*/serde_json/
 [serde]: https://docs.rs/serde/
 [std-badge]: https://badge-cache.kominick.com/badge/std-1.25.0-blue.svg
 [std]: https://doc.rust-lang.org/std