From 3c0691df1b7ff1fd04e22ab6055e84cc2137504e Mon Sep 17 00:00:00 2001 From: David Luevano Alvarado Date: Sun, 17 Apr 2022 02:13:36 -0600 Subject: refactor config handler and more Refactored the configuration handling to use configparser; also the argument parser; generally added more 'logging'; updated template resources; and more minor things --- README.md | 169 +++++++++++++++++++++++++++++++++++++++----------------------- 1 file changed, 107 insertions(+), 62 deletions(-) (limited to 'README.md') diff --git a/README.md b/README.md index 9e4275b..1a0a3b4 100644 --- a/README.md +++ b/README.md @@ -2,12 +2,8 @@ Inspired (initially) by Roman Zolotarev's [`ssg5`](https://rgz.ee/bin/ssg5) and [`rssg`](https://rgz.ee/bin/rssg), Luke Smith's [`lb` and `sup`](https://github.com/LukeSmithxyz/lb) and, pedantic.software's great (but *"mamador"*, as I would say in spanish) [`blogit`](https://pedantic.software/git/blogit/). -I'm writing this in *pYtHoN* (thought about doing it in Go, but I'm most comfortable with Python at the moment) because I want features from all of these minimal programs (and more), but I don't really want to be stitching each one of the features on any of these programs, because they're written in a way to only work as how they were first imagined to work like; I already tried adding features to `ssg` and ended up rewriting it in POSIX shell, but it was a pain in the ass when I tried to add even more, and don't get me started on trying to extend `blogit`... And also because I want to. - ## Current features -**This is still a WIP. Still doesn't build `sitemap.xml` or `rss.xml` files.** - - [x] Build static site parsing `markdown` files ( `*.md` -> `*.html`) - [x] ~~Using plain `*.html` files for templates.~~ Changed to Jinja templates. - [x] Would like to change to something more flexible and easier to manage ([`jinja`](https://jinja.palletsprojects.com/en/3.0.x/), for example). @@ -15,15 +11,20 @@ I'm writing this in *pYtHoN* (thought about doing it in Go, but I'm most comfort - [x] Tag functionality. - [ ] Open Graph (and similar) support. (Technically, this works if you add the correct metadata to the `*.md` files and use the variables available for Jinja) - [x] Build `sitemap.xml` file. + - [ ] Include manually added `*.html` files. - [x] Build `rss.xml` file. - [ ] Join the `static_url` to all relative URLs found to comply with the [RSS 2.0 spec](https://validator.w3.org/feed/docs/rss2.html) (this would be added to the parsed HTML text extracted from the MD files, so it would be available to the created `*.html` and `*.xml` files). Note that depending on the reader, it will append the URL specified in the RSS file or use the [`xml:base`](https://www.rssboard.org/news/151/relative-links) specified ([newsboat](https://newsboat.org/) parses `xml:base`). + - [ ] Include manually added `*.html` files. - [x] Only build page if `*.md` is new or updated. - [ ] Extend this to tag pages and index (right now all tags and index is built no matter if no new/updated file is present). -- [x] Configuration file as an alternative to using command line flags (configuration file options are prioritized). +- [x] Configuration file. ~~as an alternative to using command line flags (configuration file options are prioritized).~~ + - [x] Use [`configparser`](https://docs.python.org/3/library/configparser.html) instead of custom config handler. + +**Please note that I've removed the use of command line flags for now as it was too much bloat and unnecessary.** ### To be added/fixed -- [ ] Avoid the program to freak out when there are directories created in advance. +- [x] Avoid the program to freak out when there are directories created in advance. - [ ] Provide more meaningful error messages when you are missing mandatory tags in your `.md` files. ### Markdown features @@ -48,31 +49,35 @@ Just install it with `pip`: pip install pyssg ``` -*EW!*, I know..., I will try to make a PKBUILD and release it in AUR or something; hit me up if you do it to add it here. +Will add a PKBUILD (and possibly submit it to the AUR) sometime later. ## Usage -It is intended to be used as a standalone terminal program running on the "root" directory where you have the `src` and `dst` directories in (defaults for both flags). - -First initialize the directories you're going to use for the source files and destination files: +1. Get the default configuration file: ```sh -pyssg -s src_dir -d dst_dir -i +pyssg --copy-default-config -c ``` -You do not have to create any directories, in advance, the command above will do it. -Actually for the moment I will encourage you to **not create** any directories in advance. -That creates the desired directories with the basic templates that can be edited as desired (see variables available for Jinja below). Place your `*.md` files somewhere inside the source directory (`src_dir` in the command above), but outside of the `templates` directory. It accepts sub-directories. +Where `-c` is optional as by default `$XDG_CONFIG_HOME/pyssg/config.ini` is used. + +2. Edit the config file created as needed. -Strongly recommended to edit the `rss.xml` template. +- `config.ini` is parsed using Python's [`configparser`](https://docs.python.org/3/library/configparser.html), [more about the config file](#config-file). -Build the site with: +3. Initialize the directory structures (source, destination, template) and move template files: ```sh -pyssg -s src_dir -d dst_dir -t plt_dir -u https://base.url -b +pyssg -i ``` -Remember to add the mandatory meta-data keys to your `.md` files, these are: +- You can modify the basic templates as needed (see [variables available for Jinja](#available-jinja-variables)). + +- Strongly recommended to edit the `rss.xml` template. + +4. Place your `*.md` files somewhere inside the source directory. It accepts sub-directories. + +- Remember to add the mandatory meta-data keys to your `.md` files, these are: ``` title: the title of your blog entry or whatever @@ -81,54 +86,94 @@ lang: the language the entry is written on summary: a summary of the entry ``` -You can add more meta-data keys as long as it is [Python-Markdown compliant](https://python-markdown.github.io/extensions/meta_data/). +- You can add more meta-data keys as long as it is [Python-Markdown compliant](https://python-markdown.github.io/extensions/meta_data/), and these will ve [available as Jinja variables](#available-jinja-variables). + +- Also strongly recomended to add the `tags` metadata so that `pyssg` generates some nice filtering tags. -Also strongly recomended to add the `tags` test for `pyssg` to generate some nice filtering tags. +5. Build the `*.html` with: + +```sh +pyssg -b +``` + +- After this, you have ready to deploy `*.html` files. + +- For now, the building option also creates the `rss.xml` and `sitemap.xml` files based on templates, including only all converted `*.md` files (and processed tags in case of the sitemap), meaning that separate `*.html` files should be included manually in the template. + +## Config file + +All sections/options need to be compliant with the [`configparser`](https://docs.python.org/3/library/configparser.html). + +At least the sections and options given in the default config should be present: + +```ini +[path] +src=src # source +dst=dst # destination +plt=plt # template +[url] +main=https://example.com +static=https://static.example.com # used for static resources (images, js, css, etc) +default_image=/images/default.png # this will be appended to 'static' at the end +[fmt] # % needs to be escaped with another % +date=%%a, %%b %%d, %%Y @ %%H:%%M %%Z +list_date=%%b %%d +list_sep_date=%%B %%Y +[info] +title=Example site +[other] +force=False +``` + +Along with these, these extra ones will be added on runtime: + +```ini +[fmt] +rss_date=%%a, %%d %%b %%Y %%H:%%M:%%S GMT # fixed +sitemap_date=%%Y-%%m-%%d # fixed +[info] +version= # current 'pyssg' version (0.5.1.dev16, for example) +rss_run_date= # date the program was run, formatted with 'rss_date' +sitemap_run_date= # date the program was run, formatted with 'sitemap_date' +``` -So...that creates all `*.html` for the site and can be easily moved to the server. Here, the `-u` flag is technically optional in the sense that you'll not receive a warning/error, but it's used to prepend links with this URL (not strictly required everywhere), so don't ignore it; also don't include the trailing `/`. +You can add any other option/section that you can later use in the Jinja templates via the exposed config object. -For now, the `-b`uild tag also creates the `rss.xml` and `sitemap.xml` files based on templates including only all converted `*.md` files (and processed tags in case of the sitemap), meaning that separate `*.html` files should be included manually in the template. +Other requisites are: -For more options/flags just checkout `pyssg -h`. +- Urls shouldn't have the trailing slash `/`. +- The only character that needs to be escaped is `%` with another `%`. ## Available Jinja variables -Here is the list of variables that you can use specific Jinja templates with a short description. Note that all urls are without the trailing slash `/`. - -- `config` (`Configuration`) (all): configuration object containing general/global attributes, the useful ones being: - - `title` (`str`): title of the website. - - `url` (`str`): base url of the website. - - `static_url` (`str`): base static url where all static files are located, mostly needed for correct rss feed generator when using a `base` tag and using relative links to files. For more, see [](https://developer.mozilla.org/en-US/docs/Web/HTML/Element/base). - - `default_image_url` (`str`): as defined in `DEFAULT_IMAGE_URL` configuration option. - - `version` (`str`): version in numeric form, i.e. `0.5.0`. - - `run_date` (`str`): date when the program was run, with format required for rss. -- Pages: - - `all_pages` (`list(Page)`) (all): list of all the pages, sorted by creation time, reversed. - - `page` (`Page`) (`page.html`): page object that contains the following attributes: - - `title` (`str`): title of the page. - - `author` (`str`): author of the page. - - `content` (`str`): actual content of the page. - - `cdatetime` (`str`): creation datetime object of the page. - - `cdate` (`str`): formatted `cdatetime` as the configuration option `DATE_FORMAT`. - - `cdate_list` (`str`): formatted `cdatetime` as the configuration option `LIST_DATE_FORMAT`. - - `cdate_list_sep` (`str`): formatted `cdatetime` as the configuration option `LIST_SEP_DATE_FORMAT`. - - `cdate_rss` (`str`): formatted `cdatetime` as required by rss. - - `cdate_sitemap` (`str`): formatted `cdatetime` as required by sitemap. - - `mdatetime` (`str`): modification datetime object of the page. Defaults to None. - - `mdate` (`str`): formatted `mdatetime` as the configuration option `DATE_FORMAT`. Defaults to None. - - `mdate_list` (`str`): formatted `mdatetime` as the configuration option `LIST_DATE_FORMAT`. - - `mdate_list_sep` (`str`): formatted `mdatetime` as the configuration option `LIST_SEP_DATE_FORMAT`. - - `mdate_rss` (`str`): formatted `mdatetime` as required by rss. - - `mdate_sitemap` (`str`): formatted `mdatetime` as required by sitemap. - - `summary` (`str`): summary of the page, as specified in the `*.md` file. - - `lang` (`str`): page language, used for the general `html` tag `lang` attribute. - - `tags` (`list(tuple(str))`): list of tuple of tags of the page, containing the name and the url of the tag, in that order. Defaults to empty list. - - `url` (`str`): url of the page, this already includes the `config.url`. - - `image_url` (`str`): image url of the page, this already includes the `config.static_url`. Defaults to the `DEFAULT_IMAGE_URL` configuration option. - - `next/previous` (`Page`): reference to the next or previous page object (containing all these attributes). Defaults to None - - `og` (`dict(str, str)`): dict for object graph metadata. - - `meta` (`dict(str, list(str))`): meta dict as obtained from python-markdown, in case you use a meta tag not yet supported, it will be available there. -- Tags: - - `tag` (`tuple(str)`) (`tag.html`): tuple of name and url of the current tag. - - `tag_pages` (`list(Page)`) (`tag.html`): similar to `all_pages` but contains all the pages for the current tag. - - `all_tags` (`list(tuple(str))`) (all): similar to `page.tags` but contains all the tags. +These variables are exposed to use within the templates. The below list is in the form of *variable (type) (available from): description*. `section/option` describe config file section and option and `object.attribute` corresponding object and it's attribute. + +- `config` (`ConfigParser`) (all): parsed config file plus the added options internally (as described in [config file](#config-file)). +- `all_pages` (`list(Page)`) (all): list of all the pages, sorted by creation time, reversed. +- `page` (`Page`) (`page.html`): contains the following attributes (genarally these are parsed from the metadata in the `*.md` files): + - `title` (`str`): title of the page. + - `author` (`str`): author of the page. + - `content` (`str`): actual content of the page, this is the `html`. + - `cdatetime` (`str`): creation datetime object of the page. + - `cdate` (`str`): formatted `cdatetime` as the config option `fmt/date`. + - `cdate_list` (`str`): formatted `cdatetime` as the config option `fmt/list_date`. + - `cdate_list_sep` (`str`): formatted `cdatetime` as the config option `fmt/list_sep_date`. + - `cdate_rss` (`str`): formatted `cdatetime` as required by rss. + - `cdate_sitemap` (`str`): formatted `cdatetime` as required by sitemap. + - `mdatetime` (`str`): modification datetime object of the page. Defaults to `None`. + - `mdate` (`str`): formatted `mdatetime` as the config option `fmt/date`. Defaults to `None`. + - `mdate_list` (`str`): formatted `mdatetime` as the config option `fmt/list_date`. + - `mdate_list_sep` (`str`): formatted `mdatetime` as the config option `fmt/list_sep_date`. + - `mdate_rss` (`str`): formatted `mdatetime` as required by rss. + - `mdate_sitemap` (`str`): formatted `mdatetime` as required by sitemap. + - `summary` (`str`): summary of the page, as specified in the `*.md` file. + - `lang` (`str`): page language, used for the general `html` tag `lang` attribute. + - `tags` (`list(tuple(str))`): list of tuple of tags of the page, containing the name and the url of the tag, in that order. Defaults to empty list. + - `url` (`str`): url of the page, this already includes the `url/main` from config file. + - `image_url` (`str`): image url of the page, this already includes the `url/static`. Defaults to the `url/default_image` config option. + - `next/previous` (`Page`): reference to the next or previous page object (containing all these attributes). Defaults to `None`. + - `og` (`dict(str, str)`): dict for object graph metadata. + - `meta` (`dict(str, list(str))`): meta dict as obtained from python-markdown, in case you use a meta tag not yet supported, it will be available there. +- `tag` (`tuple(str)`) (`tag.html`): tuple of name and url of the current tag. +- `tag_pages` (`list(Page)`) (`tag.html`): similar to `all_pages` but contains all the pages for the current tag. +- `all_tags` (`list(tuple(str))`) (all): similar to `page.tags` but contains all the tags. -- cgit v1.2.3-70-g09d2