Use CURLOPT_WRITEFUNCTION to check the response code and content type (only allow HTML).
Also extract the title and charset during downloading chunk of data, and stop it when everything has been extracted.
Closes#579
Relates to https://github.com/shaarli/Shaarli/issues/324
Added:
- `SessionManager` class to group session-related features
- unit tests
Changed:
- `getToken()` -> `SessionManager->generateToken()`
- `tokenOk()` -> `SessionManager->checkToken()`
- inject a `$token` parameter to `PageBuilder`'s constructor
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
* translation system and unit tests
* Translations everywhere
Dont use translation merge
It is not available with PHP builtin gettext, so it would have lead to inconsistency.
With large imports it has a large impact on performances and isn't really useful.
Instead, write an IMPORT event, which let client using the history service resync its DB.
-> 15k link import done in 6 seconds.
Fixes#985
Rely on `mag_IN` (Magahi - INDIA) being unavailable when running localization
test suites, instead of `pt_BR` that is now available from Travis build images.
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
- when searching for tags you can now include '*' as wildcard placeholder
- new search reduces overall overhead when filtering for tags
- fixed combination with description tag search ('#' prefix)
- tests added
Fixes#784
From now, searching for tags with an empty value will return only not tagged links,
with the search bar showing `x results [not tagged]`.
Note that using the api, the searchtags request parameter must be set to `false` to get the same result.
- [ ] Update API doc
* The tag list can be sort alphabetically or by most used tag
* Edit/Delete are perform using AJAX, or fallback to 'do=changetag' page
* New features aren't backported to vintage theme
The goal of this is to be able to adapt the timezone form
in template without hacking the HTML already rendered.
* there are two arrays available:
* `continents` which contains only a list of available continents
* `cities` which contains a list of available timezone cities, associated with their continent
Note: there are two distinct array because RainTPL doesn't support nested loop very well.
Use case: rest API service
* saved by default in data/history
* same format as datastore.php
* traced events:
* save/edit/delete link
* change settings or plugins settings
* rename tag
Error:
An error occurred while parsing configuration JSON file (data/config.json.php): error code #4
➜ Syntax error
Please check your JSON syntax (without PHP comment tags) using a JSON lint tool such as jsonlint.com.
Use php-intl extension to display datetimes a bit more nicely, depending on the locale.
What changes:
* the day is no longer displayed
* day number and month are ordered according to the locale
* the timezone is more readable (UTC+1 instead of CET)
This setting allows to escape HTML in markdown rendering or not.
The goal behind it is to avoid XSS issue in shared instances.
More info:
* the setting is set to true by default
* it is set to false for anyone who already have the plugin enabled
(avoid breaking existing entries)
* improve the HTML sanitization when the setting is set to false - but don't consider it XSS proof
* mention the setting in the plugin README
* Private tags: make sure empty tags are properly filtered
* Search results:
* Use preg_split instead of function combination
* Add normalize_spaces to remove extra whitespaces displaying empty tags search
* REST API routes are handle by Slim.
* Every API controller go through ApiMiddleware which handles security.
* First service implemented `/info`, for tests purpose.
All existing link will keep their permalinks.
New links will have smallhash generated with date+id.
The purpose of this is to avoid collision between links due to their creation date.
* make sure we match exactly `nomarkdown` tag
* pass the whole link data to stripNoMarkdownTag() to:
* strip the noMD tag in taglist (array)
* strip the tag in tags (string)
Fixes#689
tmp
All plugins can optionally add an init function named `pluginname_init()` which is called when the plugin is loaded.
This function is aware of the config, and can return initialization errors, which are displayed in the header template.
Note that the previous error system hack no longer work.
Relates to https://github.com/shaarli/Shaarli/issues/607
Modifications:
- [application][tests] NetscapeBookmarkUtils: more permissive doctype detection
The IE bookmark exports contain extra escape sequences, which can be observed
by binary comparison of the reference input data used in tests:
$ cmp -b -l -n 8 netscape_basic.htm internet_explorer_encoding.htm
1 74 < 357 M-o
2 41 ! 273 M-;
3 104 D 277 M-?
4 117 O 74 <
5 103 C 41 !
6 124 T 104 D
7 131 Y 117 O
8 120 P 103 C
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
Relates to #607
Relates to #608
Relates to #493 (abandoned)
Additions:
- use Composer's autoload to load 3rd-party dependencies under vendor/
Modifications:
- [import] replace the current parser with a generic, stable parser
- move code to application/NetscapeBookmarkUtils
- improve status report after parsing
- [router] use the same endpoint for both bookmark upload and import dialog
- [template] update bookmark import options
- allow adding tags to all imported links
- allow selecting the visibility (privacy) of imported links
- [tests] ensure bookmarks are properly parsed and imported in the LinkDB
- reuse reference input from the parser's test data
See:
- https://github.com/shaarli/netscape-bookmark-parser
- https://getcomposer.org/doc/01-basic-usage.md#autoloading
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
* Add a new settings (which needs to be manually set): `security.trusted_proxies`
* On login failure, if the `REMOTE_ADDR` is in the trusted proxies, try to retrieve the forwarded IP in headers.
* If found, the client address is added in ipbans, else we do nothing.
Fixes#409
Relates to #102
Additions:
- application:
- export: allow prepending note permalinks with the instance's URL
- test coverage
Modifications:
- export template: switch to an HTML form
- link selection (all/private/public)
- prepend note permalinks with the instance's URL
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
* New config: `$GLOBALS['config']['REDIRECTOR_URLENCODE']` (default `true`).
* Parameter added to LinkDB constructor.
* Fixes a bug with urlencode and escaped url.
* In `index.php`, LinkDB is now instanciate once for `importFile()` and `showDaily()`.
* TU
* search type now carried by LinkDB in order to factorize code between different search sources.
* LinkDB->filter split in 3 method: filterSearch, filterHash, filterDay (we know what type of filter is needed).
* filterHash now throw a LinkNotFoundException if it doesn't exist: internal implementation choice, still displays a 404.
* Smallhash regex has been rewritten.
* Unit tests update
Going through multiple reverse proxy will store multiple scheme and port in HTTP header separated by a comma. Shaarli will use the first one to generate server_url.
* Searching '-mytag' will now exlude all shaares with 'mytag' tag.
* All tags starting with a '-' are renamed without it (through the Updater).
* Unit tests.
Minor code changes:
* LinkDB->filter() can now take no parameters (get all link depending on logged status).
* tagsStrToArray() is now static and filters blank tags.
* contains methods designed to be run once.
* is able to upgrade the datastore or the configuration.
* is based on methods names, stored in a text file with ';' separator (updates.txt).
* begins with existing function 'mergeDeprecatedConfigFile()' (options.php).
Fixes#436
Modifications:
- remove calls to strval() on safe data
- update the date format: 'Y/m/d_H:i:s' => 'Y/m/d H:i:s'
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
* `get_http_url()` renamed to `get_http_response()`.
* Use the same HTTP context to retrieve response headers and content.
* Follow HTTP 301 and 302 redirections to retrieve the title (default max 3 redirections).
* Add `LinkUtils` to extract titles and charset.
* Try to retrieve charset from HTTP headers first (new), then HTML content.
* Use mb_string to re-encode title if necessary.
* introduce class LinkFilter to handle link filter operation (and lighten LinkDB).
* handle 'private only' in filtering.
* update template to prefill search fields with current search terms.
* coding style.
* unit test (mostly move from LinkDB to LinkFilter).
PS: preparation for #358#315 and 'AND' search.
Parse link description in Markdown (HTML) before rendering.
* hard remove of Shaarli's HTML before parsing.
* Using Parsedown <https://github.com/erusev/parsedown> PHP lib.
* Includes basic markdown CSS.
* Style: removed 400px height max limit for shaares.
* Unit tests.
Fixes#328 - URL encode links when a redirector is set
* WARNING - template edit - new variable available : "real_url"
Contains the final real url (redirected or any other change on original URL)
* Don't redirect shaares link in RSS/Atom.
* Affects links shaared in description.
* Move text2clickable and keepMultipleSpaces to Utils.php + unit test
UPDATE:
* keepMultipleSpaces renamed to space2nbsp
* space2nbsp improved to handle single space at line beginning
* links in text description aren't 'nofollow' anymore
Relates to #40
Relates to #372
Additions:
- FileUtils: IOException
- ApplicationUtils:
- check if Shaarli resources are accessible with sufficient permissions
- basic test coverage
- index.php:
- check access permissions and redirect to an error page if needed:
- before running the first installation
Modifications:
- LinkDB:
- factorize datastore write code
- check if the datastore
(exists AND is writeable) OR (doesn't exist AND its parent dir is writable)
- raise an IOException if needed
Signed-off-by: VirtualTam <virtualtam@flibidi.net>
