1.3.2

Release 1.3.2 includes a few additional features detailed below, as well as fixes to all bugs discovered in releases 1.3.0 and 1.3.1. For details of significant changes and bug fixes since previous releases, view CHANGELOG.md included with the release, or the Distributed Proofreaders wiki, which also has links to the relevant sections of the Guiguts Manual.

Changes

• New auto-correction features added to error check dialog, primarily for
  Load Checkfile (to support OCRfixr output), but some features can be used
  for other error checks.
  • Ctrl+Mouse-1 makes the change suggested by the query/error (for Jeebies,
    this swaps he/be; for OCRfixr it makes the suggested correction).
  • Ctrl+Mouse-2/3 does the same, but also removes the query from the list
    (as Mouse-2/3 currently do).
  • Ctrl+Shift+Mouse-2/3 discards all queries that are identical to the
    clicked one but on a different line number. This is to quickly get rid of
    multiple wrong suggestions, and may be useful for other tools, such as
    Bookloupe. Note it does not remove all errors of that type, just the ones
    that match exactly, e.g. it can remove all occurrences of "Query digit in
    4to", but retain other digit queries.
• Regex scannos now clear the case insensitive flag by default
• The pphtml tool now outputs the line number where CSS was defined if it is
  warning about it potentially being unused.
• The Common Character dialog now has a small border to make it easier to click
  on and pop it to the front without inserting a character
• The Surround and Flood dialogs are now wider to allow them to be picked up
  and moved
• The Surround and Flood dialogs now accept input from the Common Character and
  Compose Character dialogs
• The display of match variants in the Spell Check dialog has been made clearer
  by adding additional words to the label instead of just a list of numbers,
  e.g. 3 exact, 2 case, 1 possessive, 4 hyphen in text. instead of
  3, 2, 1, 4 hyphens in text.
• Check Orphaned Brackets can now check /X X/, /F F/, /I I/ and /L L/.
• The default CSS header file has improved support for indexes, indenting
  wrapped lines more clearly
• The Preferences->Enable Quotes Highlighting option has been moved to the
  Search menu with the other highlighting options. Ctrl+semicolon toggles the
  highlighting of quotes and brackets that surround the cursor, rather than
  the previous behavior where it was necessary to edit the text to make the
  highlighting show. Also, quotes no longer need to be on the same line, and
  curly single and double quotes are supported
• A new right-click menu has been added, containing Cut, Copy and Paste options
  as well as the Bookmarks submenu. In addition, the location of the
  right-click is used for the bookmark, the Paste position, and whether to
  overwrite previously selected text
• Basic Fixup options to "Fix up spaces around hyphens" and "Format ellipses
  correctly" are now unchecked by default. "Fix up spaces around hyphens" only
  affects single hyphens, not longer dashes. "Remove space before periods"
  only affects single periods, not ellipses. Correction of ellipses is also
  more conservative in its approach, not adding a space before if there is
  one already, or if there are adjacent quotes or sentence-ending punctuation.
• Quick Count feature added to Search menu, with Shift+Ctrl+b shortcut, that
  reports the number of occurrences of the currently selected word.

Bug Fixes

• Search/Replace dialog did not remember its position correctly under Linux
• When smallcaps tools popped the Search/Replace dialog, they did not
  completely clear existing search/replace strings
• Running Stealth Scannos for a second time while the Hints dialog was popped
  would generate an error
• Check for Updates dialog did not expand to show all its contents if the font
  size was increased
• Basic Fixup checks were not skipped for /F markup as they were for /X
• An opening /X caused Basic Fixup checks to be skipped for the rest of the
  file, rather than resuming at X/
• Spacing before poetry line numbers inside /P markup was removed by Basic
  Fixup
• Scannos directory selection could generate an error if the directory name
  was too short
• Duplicate errors output by Bookloupe could cause an error when the user
  attempted to remove them both from the list
• When using Common Characters or Compose Character, an error could be
  generated if the dialog accepting input was dismissed leaving the focus in
  another dialog that could not accept that input
• The Spell Check dialog did not display the different types of matching
  (case, possessive or hyphen variants) for the first word found
• For the /P check, the Check Orphaned Brackets dialog displayed the raw regex
  string /[Pp]|[Pp]/ instead of a friendly version /P P/
• The line length check in pptext did not count multibyte characters correctly,
  such as emdashes. The shortest line determination was also incorrect
• Right-clicking the text window over 100 times would cause an error, and
  eventually the program would exit
• Checkbox backgrounds on Windows changed color instead of remaining white when
  Preferences->Appearance->Set Button Highlight Color was used, even if the
  color choosing was cancelled

1.3.1

Release 1.3.1 includes a few minor features detailed below, as well as fixes to all bugs discovered in release 1.3.0. For details of significant changes and bug fixes since previous releases, view CHANGELOG.md included with the release, or the Distributed Proofreaders wiki, which also has links to the relevant sections of the Guiguts Manual.

Changes

• New option in File-->Content Providing menu to replace tabs, curly quotes and emdashes with acceptable equivalents.
• Content-providing option to highlight characters in WF Character Count that are not in a list of enabled DP character suites. Also, ability to manage which charsuites are enabled/disabled. Characters in WF dialog can be Control-clicked to rapidly enable the appropriate character suite.
• 99% Auto page separator fixup now waits for the user to click Refresh before beginning processing.
• Guiguts and Bookloupe Test files removed from release package.
• User can now copy information from the Software Versions dialog.
• Image width type: percent, em or px is now saved between runs of the program.

Bug Fixes

• Word Frequency Check Accents was not checking Latin-1 characters
• Queries from ppvimage were not counted correctly
• If the path to an image contained parentheses, the wrong path was entered in the HTML file.
• The a shortcut to jump to words beginning with a in Word Frequency failed when there were also words beginning with æ.
• The Help->About dialog could not be resized if the font size was changed.
• Adding illustration markup to the HTML file caused the cursor to jump to the CSS section.
• Typo in regex scanno file added a space when correcting a period followed by a lowercase letter on the following line.
• An error could be caused by using Manual Smallcaps conversion when the Search dialog had only two replacement fields.
• Some files could cause the update of the status bar to take so long that the program became unresponsive.
• Some e-readers showed illustrations with widths in pixels at 100% width.
• Block wrap markup with custom margins failed to rewrap correctly if nested within no-wrap markup.
• Page numbers could be combined or skipped in HTML output if page contained only one line of main text, e.g. due to long footnote.
• Joining lines could cause a page boundary to appear mid-word, leading to pagenum spans appearing mid-word in HTML, and spaces appearing mid-word in epub versions.
• Word Frequency searching could fail when a MiXeD-CasE word had underscore markup surrounding it.
• Bookloupe View Options dialog failed to resize when font was changed.
• Error check files with line:column at start of line did not get interpreted correctly.

1.3.0

Release 1.3.0 includes all the features described in 1.3.0-beta as well as bug fixes discovered during beta testing. For details of significant changes and bug fixes since release 1.2.4, view CHANGELOG.md included with the release, or the Distributed Proofreaders wiki, which also has links to the relevant sections of the Guiguts Manual.

Changes

• Block markup can be closed and re-opened without an intervening blank line
• Betacode-style Greek Compose Sequences added using hyphen/minus
• Text entry dialogs (e.g. Compose) can be resized vertically in case user has chosen a large font
• The user can change the highlight colour used for quotes, etc., since it now uses the scannos highlight colour
• Program Files (x86) is used as an alternative default location when finding where XnView/Aspell are installed
• Kindlegen is no longer bundled with the release
• Ebookmaker 0.10.5 (current live version) is bundled with the release

Bug Fixes

• Word Frequency sorting was very slow
• Undo failed to completely undo the addition of illustration code with the Image button on the HTML Markup dialog
• Right-align numbers failed to work if the line length was longer than the rewrap margin
• The new warning about footnote mismatch (e.g. when open bracket is missing) has been removed
• 100% width illustration CSS code was sometimes duplicated
• If some View Options were hidden, bookloupe would initially report the wrong number of queries

1.3.0-beta

Information on the changes in this release is also available on the Distributed Proofreaders wiki, with links to the relevant sections of the Guiguts Manual.

Changes

• Configure Page Labels has been redesigned and can now cope with thousands of pages. Label can be selected and edited in the dialog, or by using the following shortcuts: Shift+Mouse-1 cycles Arabic/Roman/ditto; Control+Mouse-1 cycles Start@/+1/No Count; Double-click show page image. An Auto-Img button enables automatic page image display when a page is selected. Click and drag in the list to select and scroll, or use middle button drag for rapid scroll.
• Indent/Hanging Indent rewrapping is now supported in ASCII Tables. Also other minor bug fixes and improvements. Shortcut keys in the ASCII Table Effects dialog have been changed: instead of arrow keys, use the first character of Next, Previous, Left and Right.
• Straight to curly quote conversion has been added to the Txt menu and is only intended for use on text files, not HTML files. This uses an enhanced version of the ppsmq algorithm: it detects ditto marks if they have double spaces before and after; quotes at start and end of line are always selected correctly. A Curly Quote Correction submenu has been added, which can be torn off to make a mini-dialog. It allows selection of the next @ line, flipping all double quotes in selection, rechoosing double quotes based solely on spacing, removal of @ signs, selection and conversion of unconverted single quotes, and insertion of four types of quote.
• The font used in the menus, for labels, checkboxes, radio buttons, etc., can now be configured in Preferences-->Appearance-->Set Fonts. The default is to continue to use the system default. Note that it is not possible to change the font for the top menu bar.
• A fraction conversion submenu has been added to the Tools menu, allowing conversion to Unicode fractions only, superscript/subscript form, or a mixture (Unicode fraction if it exists, otherwise superscript/subscript form)
• Auto Illus Search now has the option to insert code for the current illustration and automatically load the next image file alphabetically, speeding up the insert of illustration code.
• Error checks, e.g. Bookloupe, Jeebies, Tidy, etc. have a live count of the number of queries remaining in the list displayed in the top left corner.
• Insert Page Labels (as opposed to markers) added to the Adjust Page Markers dialog and the Txt menu.
• Import and Export Prep Text Files now remember the most recent folder used
• After Import Prep Text Files, the Save As dialog is popped to allow the user to load the full file prior to beginning checks.
• User is warned if suspect footnote markup is spotted, e.g. missing opening bracket, during First Pass processing.
• Improved icon shipped with release to be used for desktop shortcuts, etc.
• 99% Auto Page Separator Fixup removes adjacent consecutive page separators
• Packaging and installation for Mac platforms improved to avoid issues running unsigned binaries. Mac users can now easily build their own Jeebies binary.
• The stickiness of Case, Regex, Whole word and Reverse can now be turned off in the Preferences menu.
• When ebookmaker is run, Guiguts now displays the busy cursor until it is completed. Any messages output by ebookmaker will be stored in the message log. If there are errors, the message log will be popped, but not if the conversion is successful.
• Check Accents in the Word Frequency dialog now includes all accented characters from the Latin Extended A and B, and Latin Extended Additional Unicode blocks.
• The Regex entry field in Word Frequency has been made wider
• Book title and author are passed to ebookmaker, so that they are included in the epub/Kindle files created.
• The Auto level used during Page Separator Fixup now defaults to 99%, and the last-used setting is saved between runs of the program.
• Wrapping of indexes enclosed in /I...I/ markup now assumes the text is formatted according to DP guidelines.
• A Highlight Alignment Column feature has been added to the Search menu, with shortcut Ctrl+Shift+a toggling it on and off. This displays a highlighted vertical line at the cursor for use when aligning frontmatter, tables, etc.
• Greek characters with and without accents and breathing can be typed with Compose Key. All Greek letters are composed with =, followed by breathing, accent, subscript, letter if applicable, e.g. Compose=a for alpha, Compose=(a for alpha with breathing, etc.
• Compose Sequence has been added to the Tools->Character Tools menu.
• Paths to tools and scannos are preserved better if the settings are copied from a previous release.
• Right-Align Numbers in Selection will now right align several numbers separated by commas, hyphens and ndashes.
• Version 0.10.3 of ebookmaker is now included with Guiguts.
• Alt-up/down text moving feature removed
• Remote W3C Validation has been removed.

Bug Fixes

• Index page numbers were not linked if the text entry ended in quotes.
• Auto-Index tried to convert some numbers to page links that it should not.
• An error was output if an ASCII table column shrank smaller than zero width.
• Scanno highlighting was not turned on when a scannos file was loaded, nor if the program was restarted, despite appearing to be enabled in the status bar.
• The line of CSS that handled overriding the image width on handheld devices was duplicated for every image with the same width dimensions.
• Cancelling Save As caused undefined variable errors.
• Several minor bugs could occur when moving footnotes inline.
• Running Auto Table HTML conversion used to lose page marker positions.
• ASCII table operations used to lose page marker positions.
• A dollar sign in matched search and replace text was sometimes deleted.
• Hyphens were not treated as word boundaries in entry fields, unlike in the main window.
• HTML page numbering was occasionally wrong when two page numbers were adjacent due to a blank page.
• Text such as I/ could be mistakenly interpreted as markup.
• The FOOTNOTES heading was not translated according to the language rc file.
• Ndashes were not treated as hyphens when converting indexes to HTML.
• Ampersand could not be included in Custom URLs.
• Guiguts lost focus when the Image Viewer was used.
• Regex replace for ^ or $ did not work.
• Footnotes without closing ] were not always flagged.
• $t in the Custom Menu couldn't handle utf-8 characters.
• Word Frequency Ligatures did not respect Suspects Only flag.
• Footnotes were not always placed correctly in chapter landing zones.
• Using Count on a regex of ^ would cause an infinite loop.
• Unhelpful error message when using non-ascii characters in filenames.
• Searching and replacing with a multi-line string terminating in a newline caused a subsequent Undo to lose text.
• Search and Replace dialog could not have its width resized.
• Bookloupe said it would not report some things and then did report them.
• If a 2-blank-line section heading inside block markup followed a page break, 99% Auto Page Separator Fixup lost one of the blank lines.
• Auto Img button could appear on when it had been turned off.
• If a page marker followed immediately after a replaced string, it would jump to the start of the string.
• Joining footnotes to previous could fail if text was edited after First Pass.
• Unicode Character Search caused an infinite loop under Perl 5.32.
• A zero-width non-breaking space or BOM could cause an error when scrolling.

1.2.4

Changes

• restored the Find Some Orphaned Markup button to the HTML Markup dialog

Bug Fixes

• illow class CSS had an unwanted space character before class name
• length and frequency sorting in Word Frequency were reversed
• error when popping Page Separator dialog on some platforms
• pasting utf8 characters into the Search & Replace dialog failed on some platforms
• orphaned brackets might not be reported and could generate error
• closing footnote divs were badly placed
• file paths with quotes caused errors when saving
• error when searching for "0" from Word Frequency dialog
• spaces at the start of a paragraph were wrongly preserved, leading to an indented first line after rewrap
• 80% auto mode in Fixup Page Separators could get stuck in a loop

1.2.3

Bug Fixes

• faulty bin file could be created, depending on page label setup
• /P...P/ poetry was not always rewrapped correctly
• search and replace dialog was too wide

1.2.2

Changes

• User can specify HTML image size in pixels, not just ems or percentage,
  by enabling the feature in the Preferences-->Processing Menu
• Post-Processing Workbench added to the Custom Menu
• Underscores are preserved when filename is used to create an image id

Bug Fixes

• error messages could be output when saving after checking curly quotes
• line numbers were hidden if rewrap was attempted on an empty selection
• selecting a word or using Ctrl+arrow keys to move by word could cause
  "Malformed UTF-8 character" errors
• pasting moderate to large amounts of text could make the program crash

1.2.1

Changes

• user can now choose which key to use to start a Compose Sequence
  via Processing in the Preferences Menu

Bug Fixes

• saving a project without page labels generated many errors
• error messages were being output too slowly
• if Aspell was not set up, Software Versions generated an error

1.2.0

Improved HTML and epub generation

• redesigned HTML Markup dialog includes buttons configurable via ctrl-click
• readability of HTML pagenum and blockquote markup is improved
• latest version of ebookmaker is bundled with the Windows release
• @media handheld is no longer used in generated HTML files
• Best Practices code is used for floated images
• chapter separator horizontal rules are hidden on paged devices

Unicode characters

• compose key (Right-Alt/ctrl+m) allows entry of characters via a
  sequence of keystrokes - list available via the Help menu
• Commonly-Used Characters chart replaces the Latin-1 chart, with spare
  buttons that the user can configure via ctrl-click
• Unicode Character Entry can be popped by right-clicking status bar
  ordinal label, and remembers previous use
• Unicode dialog is faster and more streamlined
• Greek transliteration dialog buttons show the correct character to be
  inserted in the current font, and it remembers type of input used

Usability

• using F1 key in any dialog will display the relevant manual page
• fonts may now be configured and displayed instantly for the main text window,
  Greek and Unicode dialogs, and text entry fields
• shift-clicking on search buttons temporarily reverses the search direction
• warnings and errors that used to only appear in the command window are
  saved and displayed in a message log
• word count threshold for Word Frequency's Ital/Bold/SC/etc button is saved,
  overriding default maximum of 4 words detected within those markup tags
• the Word Frequency Ital/Bold/SC/etc button also checks cite, em, strong, f, g & u
• auto-advance is now on by default for Stealth Scannos
• view options are remembered when the bookloupe dialog is closed,
  and the user's default settings are loaded when guiguts starts
• options relating to content providing are now in a submenu of the File menu
• items on the status bar have been reordered to support narrow windows

Operations sped up by factor of 3 or more

• Basic Fixup
• rewrapping
• Replace All with regular expressions
• several footnote operations
• HTML autogeneration

Removed in this version

• old menu layout
• old spellcheck layout
• old rewrap algorithm
• gutcheck tool (replaced by bookloupe)
• bookloupe run options dialog
• functionality relating to proofers' names
• debug button in multilingual spelling

Bug fixes

• Search/replace failed when text substituted for $1 contained literal $2
• goto dialogs forgot their screen position
• Greek dialog failed to resize correctly
• certain unicode strings were pasted as garbage
• image viewer was opened twice on first use
• bad text index errors occurred during sidenote fixup
• project dictionary filename could not cope with more than 9 volumes
• alphabetical sorting was incorrect in Word Frequency lists
• save and export Word Frequency lists were not utf8-safe
• a search term that was also an invalid regex caused a Replace All loop
• "Start at Beginning" setting was cleared when using Count in S&R dialog
• pphtml reported a lack of space in self-closing tag, e.g. <br/>
• Find Previous Proofer Comment failed when already at a comment
• Undo/Redo buttons were unreliable on the Page Separators dialog - the
  main undo/redo mechanism now also works for page separator changes
• converting inline sidenotes to HTML sometimes deleted text incorrectly

1.1.1

This release contains a few small improvements and bug fixes.

Changes in this version

• Variation Selectors Unicode block is now available. These combining
  characters are generally not visible, but are in order VS1 to VS16.
  Hovering over the small squares or spaces where the characters are
  should also show you which is which.
• Improvements made to Check for Updates dialog

Bug Fixes

• Bundled CSS validator reported wrong line numbers
• Draw ascii boxes failed when text was rewrapped
• Clicking in an error list window jumped to previous error's line number
• Undo/redo in Page Separator dialog sometimes output error messages
• Poetry in footnotes was enclosed in <p> markup during HTML generation
• Save My View in Bookloupe View Options output error messages
• Three or more hyphens at start/end of line not converted to HTML emdashes
• Save did not prompt for a filename if file was unnamed and not edited
• Enable/disable Autosave output error messages
• No count pages in Roman style gave errors in Configure Page Labels
• Incorrect error message displayed when running Jeebies on file without he/be