xsv 1.1.0 → 1.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f1ebfa4e4778af72a8b295d258d899a5b5d01fd029d1294d54af5e4f1e0de05a
-  data.tar.gz: aa74ffe0d57eebc12e312bdb42107bc39203cd3a0237f2e2481205c8b3b933c9
+  metadata.gz: 58e5d405e39f42d0e5287d47dd65c65b39a0ab5a2fc7fde3fd85c7211469e6e1
+  data.tar.gz: 7100a73ce192536f81a34ffbb1b431a793edf9cb71c1612a547a64a686a8330f
 SHA512:
-  metadata.gz: 6ebbb32e48860043bdb0a5d17f6fef525252e8bb01ac63180cd0e749dfbd1b3bb08b8a82e43e9e13da2fa187c02f77e37f525e9c7453334bc3cb8fdf05400187
-  data.tar.gz: 4e1450daebcc3ddfbc0585de52f4d1f362ef2d79281e59cc641119a47924f8450b76c3643a6403a440e0a581ebfcc108431a311f902e2b48be61a1d2afd7b19e
+  metadata.gz: 3b8fcbab2e2aa1f02dc0b51051a9b60dd2518b18b72007c2f3e77fa99248e864069d54b0bb43d783f8bb6ef79b6c2504c8cd05c244a2e9c85cddb882de224556
+  data.tar.gz: 3ec5120d8b6e365996985c75f4c291e3f4805e9876fd93ae2dfe071c5bd69ad751677cf71121ac23f6e9bed75ab80296b12346d79696cb6105974345d289bb7e

data/CHANGELOG.md

@@ -1,5 +1,17 @@
 # Xsv Changelog
 
+## 1.2.0 2023-01-01
+
+**This release contains the following minor breaking changes**
+
+- Raise an error when entering hash mode on a sheet with duplicate headers to prevent unintentional behaviour (fixes #44)
+- Xsv now returns frozen strings to further improve performance. This means it's no longer possible to call mutating methods on strings read from worksheets without unfreezing them first.
+- Unescape all HTML entities in XML characters (thanks @til)
+
+## 1.1.1 2022-04-01
+
+- Improve compatibility with files generated by the Open XML SDK (#40)
+
 ## 1.1.0 2022-02-13
 
 - New, shorter `Xsv.open` syntax as a drop-in replacement for `Xsv::Workbook.open`, which is still supported
@@ -115,4 +127,4 @@ Fix a Gemfile small Gemfile issue that broke the 0.3.3 and 0.3.4 releases
 
 ## 0.3.3 - 2020-03-02
 
-Intial version with a changelog and reasonably complete YARD documentation.
+Initial version with a changelog and reasonably complete YARD documentation.

data/README.md

@@ -1,9 +1,11 @@
 # Xsv .xlsx reader
 
-[![Travis CI](https://img.shields.io/travis/martijn/xsv/master)](https://travis-ci.org/martijn/xsv)
-[![Codecov](https://img.shields.io/codecov/c/github/martijn/xsv/main)](https://app.codecov.io/gh/martijn/xsv)
-[![Yard Docs](http://img.shields.io/badge/yard-docs-blue.svg)](https://rubydoc.info/github/martijn/xsv)
-[![Gem Version](https://badge.fury.io/rb/xsv.svg)](https://badge.fury.io/rb/xsv)
+
+
+[![Test badge](https://img.shields.io/github/actions/workflow/status/martijn/xsv/ruby.yml?branch=main)](https://github.com/martijn/xsv/actions/workflows/ruby.yml)
+[![Codecov badge](https://img.shields.io/codecov/c/github/martijn/xsv/main)](https://app.codecov.io/gh/martijn/xsv)
+[![Yard Docs badge](http://img.shields.io/badge/yard-docs-blue.svg)](https://rubydoc.info/github/martijn/xsv)
+[![Gem Version badge](https://badge.fury.io/rb/xsv.svg)](https://badge.fury.io/rb/xsv)
 
 Xsv is a fast, lightweight, pure Ruby parser for ISO/IEC 29500 Office Open XML spreadsheet files
 (commonly known as Excel or .xlsx files). It strives to be minimal in the
@@ -35,10 +37,9 @@ Or install it yourself as:
 
     $ gem install xsv
 
-Xsv targets ruby >= 2.5 and has a just single dependency, `rubyzip`. It has been
-tested successfully with MRI, JRuby, and TruffleRuby. Due to the lack of
-native extensions should work well in multi-threaded environments or in Ractor
-when that becomes stable.
+Xsv targets ruby >= 2.6 and has a just single dependency, `rubyzip`. It has been
+tested successfully with MRI, JRuby, and TruffleRuby. It has no native extensions
+and is designed to be thread-safe.
 
 ## Usage
 
@@ -84,8 +85,11 @@ sheet.parse_headers!
 sheet[0] # => {"header1" => "value1", "header2" => "value2"}
 ```
 
-Be aware that hash mode will lead to unpredictable results if the worksheet
-has multiple columns with the same header. `Xsv::Sheet` implements `Enumerable` so along with `#each`
+Because of the way Ruby hashes work will raise `Xsv::DuplicateHeaders` if it detects
+duplicate values in the header row when calling `#parse_headers!` or when opening
+a workbook with `parse_headers: true`.
+
+`Xsv::Sheet` implements `Enumerable` so along with `#each`
 you can call methods like `#first`, `#filter`/`#select`, and `#map` on it.
 
 ### Opening a string or buffer instead of filename

data/lib/xsv/sax_parser.rb

@@ -1,8 +1,10 @@
 # frozen_string_literal: true
 
+require "cgi"
+
 module Xsv
   class SaxParser
-    ATTR_REGEX = /((\S+)="(.*?)")/m
+    ATTR_REGEX = /((\p{Alnum}+)="(.*?)")/mn
 
     def parse(io)
       responds_to_end_element = respond_to?(:end_element)
@@ -36,14 +38,7 @@ module Xsv
             chars = pbuf.slice!(0, o + 1).chop!.force_encoding("utf-8")
 
             if responds_to_characters && !chars.empty?
-              if chars.index("&")
-                chars.gsub!("&", "&")
-                chars.gsub!("'", "'")
-                chars.gsub!(">", ">")
-                chars.gsub!("&lt;", "<")
-                chars.gsub!("&quot;", '"')
-              end
-              characters(chars)
+              characters(CGI.unescapeHTML(chars))
             end
 
             state = :look_end
@@ -67,13 +62,15 @@ module Xsv
               args = nil
             end
 
+            stripped_tag_name = strip_namespace(tag_name)
+
             if tag_name.start_with?("/")
-              end_element(tag_name[1..]) if responds_to_end_element
+              end_element(strip_namespace(tag_name[1..])) if responds_to_end_element
             elsif args.nil?
-              start_element(tag_name, nil)
+              start_element(stripped_tag_name, nil)
             else
-              start_element(tag_name, args.scan(ATTR_REGEX).each_with_object({}) { |m, h| h[m[1].to_sym] = m[2] })
-              end_element(tag_name) if responds_to_end_element && args.end_with?("/")
+              start_element(stripped_tag_name, args.scan(ATTR_REGEX).each_with_object({}) { |(_, k, v), h| h[k.to_sym] = v })
+              end_element(stripped_tag_name) if responds_to_end_element && args.end_with?("/")
             end
 
             state = :look_start
@@ -85,5 +82,16 @@ module Xsv
         end
       end
     end
+
+    private
+
+    # I am not proud of this, but there's simply no need to deal with xmlns for this application ¯\_(ツ)_/¯
+    def strip_namespace(tag)
+      if (offset = tag.index(":"))
+        tag[offset + 1..]
+      else
+        tag
+      end
+    end
   end
 end

data/lib/xsv/shared_strings_parser.rb

@@ -6,7 +6,7 @@ module Xsv
   class SharedStringsParser < SaxParser
     def self.parse(io)
       strings = []
-      new { |s| strings << s }.parse(io)
+      new { |s| strings << -s }.parse(io)
       strings
     end
 

data/lib/xsv/sheet.rb

@@ -83,6 +83,12 @@ module Xsv
     # @return [self]
     def parse_headers!
       @headers = parse_headers
+
+      # Check for duplicate headers, but don't care about nil columns
+      if (duplicate_header = @headers.detect { |h| @headers.count(h) > 1 })
+        raise Xsv::DuplicateHeaders, "Duplicate header '#{duplicate_header}' found, consider parsing this sheet in array mode."
+      end
+
       @mode = :hash
 
       self

data/lib/xsv/sheet_rows_handler.rb

@@ -85,7 +85,7 @@ module Xsv
       when "s"
         @workbook.shared_strings[@current_value.to_i]
       when "str", "inlineStr"
-        @current_value.strip
+        -@current_value.strip
       when "e" # N/A
         nil
       when nil, "n"

data/lib/xsv/sheets_ids_handler.rb

@@ -17,7 +17,7 @@ module Xsv
     end
 
     def start_element(name, attrs)
-      @block.call(attrs.slice(:name, :sheetId, :state, :'r:id')) if name == "sheet"
+      @block.call(attrs.slice(:name, :sheetId, :state, :id)) if name == "sheet"
     end
   end
 end

data/lib/xsv/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Xsv
-  VERSION = "1.1.0"
+  VERSION = "1.2.0"
 end

data/lib/xsv/workbook.rb

@@ -93,8 +93,11 @@ module Xsv
       @zip.glob("xl/worksheets/sheet*.xml").sort do |a, b|
         a.name[/\d+/].to_i <=> b.name[/\d+/].to_i
       end.map do |entry|
-        rel = @relationships.detect { |r| entry.name.end_with?(r[:Target]) && r[:Type].end_with?("worksheet") }
-        sheet_ids = @sheet_ids.detect { |i| i[:"r:id"] == rel[:Id] }
+        rel = @relationships.detect do |r|
+          entry.name.end_with?(r[:Target].sub(/^\//, "")) && # ignore leading / in some files
+            r[:Type].end_with?("worksheet")
+        end
+        sheet_ids = @sheet_ids.detect { |i| i[:id] == rel[:Id] }
         Xsv::Sheet.new(self, entry.get_input_stream, entry.size, sheet_ids).tap do |sheet|
           sheet.parse_headers! if mode == :hash
         end

data/lib/xsv.rb

@@ -21,6 +21,8 @@ require "xsv/workbook"
 module Xsv
   class Error < StandardError; end
 
+  class DuplicateHeaders < StandardError; end
+
   # An AssertionFailed error indicates an unexpected condition, meaning a bug
   # or misinterpreted .xlsx document
   class AssertionFailed < StandardError; end

data/xsv.gemspec

@@ -21,7 +21,7 @@ Gem::Specification.new do |spec|
   if spec.respond_to?(:metadata)
     spec.metadata["homepage_uri"] = spec.homepage
     spec.metadata["source_code_uri"] = "https://github.com/martijn/xsv"
-    spec.metadata["changelog_uri"] = "https://github.com/martijn/xsv/CHANGELOG.md"
+    spec.metadata["changelog_uri"] = "https://raw.githubusercontent.com/martijn/xsv/main/CHANGELOG.md"
   else
     raise "RubyGems 2.0 or newer is required to protect against " \
           "public gem pushes."

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: xsv
 version: !ruby/object:Gem::Version
-  version: 1.1.0
+  version: 1.2.0
 platform: ruby
 authors:
 - Martijn Storck
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2022-02-13 00:00:00.000000000 Z
+date: 2023-01-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rubyzip
@@ -141,7 +141,7 @@ licenses:
 metadata:
   homepage_uri: https://github.com/martijn/xsv
   source_code_uri: https://github.com/martijn/xsv
-  changelog_uri: https://github.com/martijn/xsv/CHANGELOG.md
+  changelog_uri: https://raw.githubusercontent.com/martijn/xsv/main/CHANGELOG.md
 post_install_message:
 rdoc_options: []
 require_paths: