Merge branch 'release/alpha/master'

Author: coderarjob

CMakeLists.txt

@@ -0,0 +1,67 @@
+cmake_minimum_required(VERSION 3.10 FATAL_ERROR)
+project(kpdfsync LANGUAGES NONE)
+
+find_package(Java 1.8 REQUIRED)
+include(cmake/GetVersion.cmake)
+include(cmake/AddResource.cmake)
+include(cmake/JavaCmake.cmake)
+
+set(KPDFSYNC_BIN_DIR ${CMAKE_BINARY_DIR}/bin)
+set(KPDFSYNC_LIBS_DIR ${CMAKE_CURRENT_SOURCE_DIR}/lib)
+set(KPDFSYNC_DOCS_DIR ${CMAKE_CURRENT_SOURCE_DIR}/docs)
+set(KPDFSYNC_RELEASE_DIR ${CMAKE_CURRENT_SOURCE_DIR}/release)
+set(KPDFSYNC_PACK_DIR ${KPDFSYNC_RELEASE_DIR}/current)
+set(KPDFSYNC_RELEASE_RESOURCE_DIR ${KPDFSYNC_RELEASE_DIR}/resource)
+
+set(KPDFSYNC_JDK_VER_TARGET 8)
+set(KPDFSYNC_VER 0.0.0   CACHE STRING "Application version.")
+set(KPDFSYNC_TAG <undef> CACHE STRING "Application version tag.")
+
+# Used by cmake/JavaCmake.cmake
+set(CMAKE_JAVA_COMPILE_FLAGS -target ${KPDFSYNC_JDK_VER_TARGET}
+                             -source ${KPDFSYNC_JDK_VER_TARGET})
+set(JAVA_CLASS_DIR ${CMAKE_BINARY_DIR}/class)
+
+set (KPDFSYNC_PACK_DIR_FILES
+    ${KPDFSYNC_RELEASE_RESOURCE_DIR}/kpdfsync.bat
+    ${KPDFSYNC_RELEASE_RESOURCE_DIR}/kpdfsync
+    ${KPDFSYNC_DOCS_DIR}/QuickReference.pdf
+    ${CMAKE_CURRENT_SOURCE_DIR}/LICENSE
+    ${CMAKE_CURRENT_SOURCE_DIR}/README.md
+)
+
+add_custom_command(
+    OUTPUT
+        ${KPDFSYNC_BIN_DIR}/tools
+        ${KPDFSYNC_BIN_DIR}/pdfclown.jar
+    COMMAND ${CMAKE_COMMAND} -E copy_directory ${KPDFSYNC_RELEASE_RESOURCE_DIR}/tools
+                                               ${KPDFSYNC_BIN_DIR}/tools
+    COMMAND ${CMAKE_COMMAND} -E copy ${KPDFSYNC_LIBS_DIR}/pdfclown.jar ${KPDFSYNC_BIN_DIR}
+)
+
+add_custom_target(build-all ALL)
+
+add_custom_target(mv-deps
+    DEPENDS ${KPDFSYNC_BIN_DIR}/tools
+            ${KPDFSYNC_BIN_DIR}/pdfclown.jar)
+
+add_custom_target(run
+    COMMAND ${CMAKE_COMMAND} -E env java -jar coderarjob.kpdfsync.jar
+    WORKING_DIRECTORY ${KPDFSYNC_BIN_DIR}
+)
+
+add_custom_target(pack
+   COMMAND ${CMAKE_COMMAND} -E remove_directory ${KPDFSYNC_PACK_DIR}
+   COMMAND ${CMAKE_COMMAND} -E make_directory ${KPDFSYNC_PACK_DIR}
+   COMMAND ${CMAKE_COMMAND} -E copy           ${KPDFSYNC_PACK_DIR_FILES} ${KPDFSYNC_PACK_DIR}
+   COMMAND ${CMAKE_COMMAND} -E copy_directory ${KPDFSYNC_BIN_DIR} ${KPDFSYNC_PACK_DIR}/bin
+   COMMAND ${CMAKE_COMMAND} -E echo "Package created in directory: ${KPDFSYNC_PACK_DIR}"
+)
+
+add_dependencies (pack mv-deps build-all)
+add_dependencies(build-all coderarjob.ajl coderarjob.kpdfsync.lib coderarjob.kpdfsync)
+add_dependencies(run build-all mv-deps)
+
+add_subdirectory(src/coderarjob/ajl)
+add_subdirectory(src/coderarjob/kpdfsync/lib)
+add_subdirectory(src/coderarjob/kpdfsync/poc)

README.md

@@ -46,25 +46,62 @@ Here is the rough roadmap of the development.
 
 ----
 
+- [X] Bug Fixes: Some minor fixes in UI
+- [X] Feature: Added new parser for reading older Kindle Clippings format.
+- [X] UI change to select different parsers.
+- [ ] **Alpha Release 4**
+
+----
+
 - [ ] Fine tune the rough edges in the supporting library.
 - [ ] Memory/Resource optimization.
 - [ ] Finalizing and optimizing the Graphical User Interface.
 - [ ] **Beta Release**
 
-## Requirements
-- JRE 1.8 or higher
-- Linux, Windows
-- Mac OS
+----
+
+## Installation
+
+Once you have downloaded the [release binaries](https://github.com/coderarjob/kpdfsync/releases), and the minimum requirements are met,
+to start Kpdfsync, execute the `kpdfsync` file in Linux/Mac OS or `kpdfsync.bat` in Windows.
+
+### Minimum requirements
+- JRE 1.8 __Note: JRE headless will not work.__
+- Linux, Windows 7, Mac OS Sierra(\*)
+- Linux: poppler-utils(\*\*)
+- Windows: poppler-utils(\*\*) is included with the release. (will work out of the box)
+
+(\*) PDF fixing feature does not work on Mac OS.
+(\*\*) Required only for PDF fixing feature.
+
+----
 
-**Note:** PDF fixing feature is not available on the Mac.
+## Building from source
 
-#### For fixing feature to work, you require the following:
-* Linux: poppler-utils
-* Windows: poppler-utils is included with the release.
-* Mac OS: __Not supported__
+### Minimum requirements
+- Linux, Windows 7, Mac OS Sierra
+- JDK 8
+- CMake 3.10
 
-**Note:** This is a new feature, which can resolve few of the PDF errors, but not all. New fixes
-will be added later on.
+### Steps
+
+Clone the repo or download the source. Then to build, run the following command at the top-level of
+the source directory.
+
+This will build jar files in `build/bin` directory, then run `build/bin/coderarjob.kpdfsync.jar`.
+
+```
+$ cmake -Bbuild
+$ cd build
+$ cmake --build . run
+```
+
+----
+
+## Troubleshooting
+
+* (Windows) Error: api-ms-win-crt-runtime-l1-1-0.dll is missing.
+Install (Update for Universal C Runtime)[https://support.microsoft.com/en-us/kb/2999226]
 
 ## Troubleshooting
 
@@ -76,3 +113,5 @@ Install [Update for Universal C Runtime](https://support.microsoft.com/en-us/kb/
 * PDF Clown library is used to read and highlight on PDF files. PDF Clown library is covered under
 LGPL (GNU Lesser General Public License).
 More information and source code can be found here [http://www.pdfclown.org](http://www.pdfclown.org)
+
+* Poppler is covered under GPL3 license.

TODO

@@ -28,6 +28,75 @@ Kpdfsync
           ByteOrderMarkTypes enum. It is now separe in
           ByteOrderMark file.
 # BUGS:
+[X] (GitHub Issue #9)
+    Book: Steve Jobs (MOBI)
+    Highlights and notes are not getting automatically mapped
+
+    On some pages, getting the yellow exclamation point,
+    indicating that there are nore notes than highlight.
+    However that is not the case in the Clippings file.
+
+Cause:
+    HighlightNotePairManager.pairAutomatic() loops though each PageResource
+    and calls pairAutomatic on it. This loop stops at the 1st error, so if
+    a pairAutomatic call fails, all subsequent page resources are not paired
+    automatically.
+    This came to light due to the below `#8` bug. Because that exception
+    is thrown before, no page resource after that is paired automatically.
+
+Solution:
+    1. Do not break the loop on exception. Skip the page resource which
+    cannot be mapped and go to the next. The message box with the error
+    will not appear at that point, but the `Bug` icon will be there.
+
+    2. Have a method in PageResource to dry run automatic pairing, if
+    this function returns false, skep it.
+
+[ ] (GitHub Issue #8)
+    Book: Steve Jobs (MOBI)
+    Invalid. There are more notes than highlights
+
+    Invalid. There are more notes than highlights is wrongly
+    reported even when there is no issue in the Clippings.txt
+    file.
+
+Cause:
+    On multi page highlights, there are two location numbers in
+    the clippings file. If there is also a note associated with
+    the highlight, it however has only one page in the clippings
+    file.
+    Say the highlights are on page 1107-1108 and the note is on
+    1108. Kpdfsync will parse and read the 1st location number
+    of the highlight and the only location number of the note.
+    Which means, highlight location will be 1107 and note
+    locaiton will be 1108. Thus the error - A single note in a
+    page.
+
+    Note that, if there was one highligh in 1108-1108 page, then
+    the user will see no error message, but 1108 will have note
+    whose assiciated highlight is in page 1107. See solution.
+
+    [Kindle does not do multipage highlights on PDF files, so this
+    issue not possible in PDF files]
+
+Solution:
+    1. When double clicking on a highlight, allow notes from other
+    pages to be associated with - in which case move the note to the
+    page of the highlight.
+
+    2. When clicking a page which more notes than highlights, we get
+    an error. After the error, show the NotesHighlights Map dialog and
+    which will show the notes in the page, then allow selected notes
+    to be moved to a different page.
+
+    Both this options will move the note to a different page and this
+    solving the error. Option 1 starts which the highlight, option 2
+    starts from the note, thats all the difference. In both cases the
+    user need to know the page of the offending note or the correct
+    page where the associated highlight resides.
+    I prefer the 1st option, because it is easy to know the page of the
+    offending note.
+
 [ ] The string matching algo is too simple, and gives wrong match
     percentage, if the strings being compared differ in the number
     of non-whitespace characters. The two indexes get out of sync
@@ -277,4 +346,8 @@ java.io.EOFException
 
     Solution:
     Modifed pdfclown to treat 'name' and 'post' tables as optional. It is 
-    released with kpdfsync 0.8.2-alpha.
+    released with kpdfsync 0.9.0-alpha.
+
+[ ] Book: Rust Programming Language (Duplicate issue)
+    Highlight is not visible on the output PDF file. The 'Annotations' list shows that the
+    highlights and comments exits (comments contents match) but are not visible.