The way in which the external libraries are added to the project is quite messy currently. For example one needs nlohmann_json as well as GoogleTest to be installed, but CLI11 is integrated via the CMake FetchContent feature. It would be nice if this could be unified, preferable via FetchContent maybe.

The program version has currently to be maintained in the source code as well as in the CMakeLists.txt file. This should be changed in a way, that the version is taken fom CMakeLists.txt.

Is your feature request related to a problem? Please describe.
One might want to export (part of) the library to use it for other purposes. Generating (HTML) reports would be nice to have an overview over a library.

Currently there is only one field for an author implemented as a std::string. One might add a vector of authors to enable more flexible filtering and statistics.

Is your feature request related to a problem? Please describe.
Although passing commandline arguments for adding a book is nice for scripting, sometimes it might be more user friendly and comfortable to have an interactive mode to add information.

Describe the solution you'd like
A call like "booktrack-cli -l bib.json repl" should start an interactive mode in which the user is asked to enter the information about a book.

Describe alternatives you've considered
The interactive mode could be also available for editing books and all the other subcommands. Also it might be possible to start the interactive mode automatically when no commandline argument is given or something like that.

Is your feature request related to a problem? Please describe.
Currently the list subcommand simply returns a list of all books included in the given library. The user should have the possibility to add filters to the list subcommand to get a better overview over his or her library.

Describe alternatives you've considered
It would be possible to add an own subcommand but I guess this would be not that user-friendly.

The complete build progress and testing is currently only done inside a relatively new Linux distribution. It would be expand the user base if the program is usable under Windows.

Describe the solution you'd like
There is a quite popular mobile App called Bookshelf. It would be nice to have an import function for booktrack-cli.

Add another subcommand import maybe with some parameters for diferrent formats. It has to be decided what to do with duplicates.

Is your feature request related to a problem? Please describe.
Currently the book id is more or less increased for every added book. If one deletes a book entry right in the middle of the library, there will be a gap in the series of book IDs.

Describe the solution you'd like
There should be some automatic "houskeeping" to fill that gaps. This might be triggered on a certain amount of library items maybe.

Describe the bug
When adding a book without reading data the Statistics page shows "1970" in the "Books finished by year" section.

To Reproduce
Steps to reproduce the behavior:

1. Add a book without reading data (booktrack-cli add [...])
2. show the statistics of the library (booktrack-cli statistics)

Expected behavior

1. only finished books should appear in the "Books finished by year" section
2. unfinished (or started) books should appear in a separate section or as a new entry in the "Library Statistics Summary"

Version information:

• OS: Xubuntu 20.04.6 LTS
• program version v0.3.0

Is your feature request related to a problem? Please describe.
In some cases it might be convenient to import a bulk of books via a single (sub)command.

Is your feature request related to a problem? Please describe.
Currently the use of cppcheck or similar tools is a manual step. There should be a more easy way to use this tool. At least a documentation should be provided.

Is your feature request related to a problem? Please describe.
This is a big and important feature: the possibility to edit already stored books in a library is completly missing at the moment.

Describe the solution you'd like
There might be another subcommand edit to choose a book.

Describe alternatives you've considered
...

Additional context
There might be some internal changes because currently there is no unique book id stored which could be used to choose the library entry to edit.

Is your feature request related to a problem? Please describe.
It is kind of inconvenient to completly write out every command, subcommand and option.

Describe the solution you'd like
At least the Bash does have a autocomplete feature, other shells should do too. There should be a Bash completion script or a more platform independent soulution to simplify the use of booktrack-cli.

Describe alternatives you've considered
None, yet.

Is your feature request related to a problem? Please describe.
Currently the only way to see the information about a book is using the list subcommand. Nevertheless, the horizontal space on a screen is limited and it is not possible or at least confusing to list all data there. So as a user I would like to have a new view for showing all the details of a specific book.

Describe the solution you'd like
Adding a new subcommand like info --id to view the details of a book should be a suitable first step.

Additional context
A problem might be, that the book id has to be known and therefore searched with the list subcommand first. Also one might want to see the details of more than one book at once.

Is your feature request related to a problem? Please describe.
When trying to reconstruct a bug it is helpful to have data about the runtime environment of the system, software version of booktrack-cli and so on. There should be an option to get this data for the user.

Describe the solution you'd like
An additional option for the command line should be sufficient, e.g. booktrack-cli --diagnosis.

It would be nice to have an own logo for the program to improve the recognition value.

Is your feature request related to a problem? Please describe.
There should be at least statistics like "finished books per year" and "read pages per year".

Describe the solution you'd like
Maybe another subcommant stats or statistics.

Is your feature request related to a problem? Please describe.
There is often a mismatch with the capitalization of book entries. For example sometimes all letters are in lowercase, some are uppercase and so on. This might distract the reader of a book list. It would be nice to have a method to automatically transform the texts and normalize them in a way.

Describe the solution you'd like
A new subcommand modify or transform might be implemented. In a first step it might be sufficient to add options like --title-uppercase and similar. Another step could be to introduce templates for this.

Is your feature request related to a problem? Please describe.
There is currently no explicit validation if the used library file is correct. This could lead to undefined behaviour and errors when opening such a file.

Describe the solution you'd like
Maybe using a JSON Schema for this purpose would be a benefit.

The source code documentation is not complete at the moment. There should be Doxygen compatible comments in the header files to ease the simplify the use of the classes and methods.

1. A large part of the comanndline parsing is done in main.cc currently, like the implementation of the callbacks for sub commands. This should be reorganized. The CLI11 project has an example for this.
2. delete subcommand: the choice which option was passed by this subcommand (e.g. --by-title or --by-id) is exclusive. However, this is checked by the value which is included by the option.del struct. It should be checked if CLI11 has an option to check if a parameter was passed.

Some meta information like the a custom name could be useful for the library handling.

Describe the bug
setw does not work with umlauts very well, so the whole library "dump" to std::cout should be refactored. Also it is not well located in the Library class.

Describe the bug
When manually editing the library file and changing the JSON scheme the program fails reading and crashes.

To Reproduce
Steps to reproduce the behavior:

1. Manually edit the library file by removing the "reading" entry for example
2. load the library file by using the list subcommand for example
3. booktrack-cli crashes

Expected behavior
To be defined:

1. ignore missing entries with optionally give a warning
2. Cancel the reading when the JSON scheme does not fit (see also #9)

Version information:

• OS: Debian 11 "Bullseye"
• program version: v0.4.0

It would be nice to have a report of the code coverage as a result of the implemented unit tests.

Is your feature request related to a problem? Please describe.
To be independent of the commandline interface there should be another way to access the tracked books in a library. This opens also the possibility for a new project like a Web UI.

Is your feature request related to a problem? Please describe.
Entering reading dates currently has to have the format [day].[month].[year], that is, the separator has to be a dot. This is only reasonable for german speaking countrys (?).

Describe the solution you'd like
The minimum feature set should be to accept also other separators. But actually the dates should be entered as the user is used to it.

Is your feature request related to a problem? Please describe.
In some cases it might be comfortable to have a complete build environmant in a Docker image. So the developer does not have to handle all dependencies on his or her own.

Is your feature request related to a problem? Please describe.
Currently there is no difference between a missing library (not intended) and a library that is cannot be opened for some reason. So at least at the first start there could be a little bit more information for the user.

Currently there is only one field for a title. This could be seperated into a main and a subtitle of a book for more flexible filtering and statistics.

Is your feature request related to a problem? Please describe.
The ISBN is a major feature for managing book libraries. So it should be possible to store it optionally for every book.

Describe the solution you'd like
When adding a book there should be an additional parameter for entering the ISBN of the book. There might be a function to check the validity of the entered number.

Is your feature request related to a problem? Please describe.
Currently the use of valgrind is a manual step. There should be a more easy way to use this tool. At least a documentation should be provided.

Is your feature request related to a problem? Please describe.
When adding a new book which already exists in the library, there is no message to the user.

Describe the solution you'd like
It should be checked if the book is already in the library.

Additional context
It should be evaluated and documentadet which book features are considered when doing the check (e.g. pages can definitly be the same for several books).

Currently the CI pipeline for Pull requests does not observes the state of unit tests. This should be added.

Is your feature request related to a problem? Please describe.
Sometimes I am not interested in the whole library content but only in the books I am currently reading or already read. This should be an option to filter for.

Describe the solution you'd like
Adding a flag --finished and --reading to the list subcommand could solve this.

Describe alternatives you've considered
A workaround would be to add the books to different shelfs. Filtering for this is already working, but because of the missing possibility to edit an existing book, this is not acceptable.

Is your feature request related to a problem? Please describe.
To get started working on this project it would be a relief for developers to have a "big picture" of the software architecture. Although the source code is not that large at the moment, it will become more important later on.

Describe the solution you'd like
It should be possible to draw some charts using Mermaid in the repository markdown files. It might be useful to add a DEVELOPER.md.

Is your feature request related to a problem? Please describe.
Windows is not used and tested currently so just drop the GitHub action for now.

Is your feature request related to a problem? Please describe.
Currently books can only be deleted by the title. This should be much more flexible.