I am writing a number of Google Packaged Apps which run independently, but share lots of code. For example, they all use "library.js". I would like to have only one copy of library.js so any changes to it will be used by all newly packed apps.
To package my apps, it seems they all must have a copy of library.js in their own directory structure, whereas it would be nice to have a single master copy in some other directory that is accessible to all. I currently do a manual check to make sure all files are up-to-date before packing, and I am writing some code to do the check automatically, but it seems like a work-around.
Can a Google Packaged App use JS code in external library directories, or must all code be under the root directory of the app (i.e., requiring copying from external directory) when packing?
Have you tried providing a URL i.e. host the javscript file in .js format to an accessible location to your apps and then provide the .js file URL in all your apps code. The very next time you want to change, all you have to do is to update that .js file.
Related
Pyscript allows one to run python inside a web browser. I have two python scripts I wrote that I’d like to use. One way to do this is to copy and paste the python code held in these files directly into the index.html file where the index file is part of a GitHub.io page. If possible however, I would rather load/Import them from a remote location. Currently, they reside in the gh-page branch on GitHub alongside the index.html file.
My question is whether this is possible? Most tutorials show how to load and import a local python file which I don’t want to do.
Update: This is my current attempt which I add to the index.html file:
<py-config>
[[fetch]]
from = "https://github.com/etc/blob/gh-pages/"
files = ["myadd.py"]
</py-config>
When I try this I get the error message:
(PY0001): PyScript: Access to local files (using "Paths:" in ) is not available when directly opening a HTML file; you must use a webserver to serve the additional files. See this reference on starting a simple webserver with Python.
I want to avoid starting a server because this is meant to be client-side only approach with only a dumb file repo at the other end.
There is a solution, and it's very simple, just use the syntax:
<py-script src="mypythonscript.py"> </py-script>
And it will pick up the file from the GitHub directory.
I'm learning Aurelia via the TypeScript / ASP.NET Core skeleton navigation app. Everything runs fine in its default state. To test the exported production bundle, I run the Gulp Export task, then publish the app via Visual Studio project publish to a local folder, then replace the published wwwroot folder with the wwwroot folder from within the "export" folder, then use dotnet from the command line to run the app.
Things break if I have any source modules in a subdirectory. For example, I moved the welcome.ts/html component files into "/src/Pages" and adjusted its route moduleId in app.ts accordingly to "./pages/welcome". The unbundled app then still runs up fine, but when I try the exported version, I get a request being made by SystemJS to http://localhost:5000/dist/pages/welcome.js which 404s (as you'd expect).
I can see the contents of the welcome component in the app-build.js file, and the config.js file within the export folder contains the expected file paths, i.e. it has "Pages/welcome...".
I have read this seemingly similar issue:
https://github.com/aurelia/bundler/issues/131
But setting depCache to false made no difference in my case. Why is SystemJS trying to load this module separately from outside of the bundle?
I was able to reproduce this error locally.
Presuming that you have a Windows environment, it will be a case-sensitivity issue.
After renaming [P]ages folder to [p]ages, bundled version works as expected.
On the filesystem there is a [P]ages/welcome.js viewmodel, but [p]ages/welcome has been defined as moduleId.
Unbundled mode: Windows filesystem is case-insensitive, which behaviour can be misleading by loading [P]ages/welcome.js correctly.
Bundled mode:
Based on file path, bundling process embeds[P]ages/welcome.js as [P]ages/welcome module.
But, according to the route config, SystemJS will be looking for [p]ages/welcome module within app-build.js.
My recommendation would be to use lowercase folder/filenames whenever it's possible.
I want to deploy some front end assets to the local web root of a site using file replacement. I can't seem to get it to work with a relative path in the target location field though. Is it possible to do this though tds or should I use a post build event instead?
The reason these assets aren't included in a project is that they are part of a third party solution but we still want this tracked in source control to try to make the project setup easier.
Most developer machines will be set up the same way for this project with the same file structure but I think it's a little more flexible if I can make the target a relative path so I don't need to worry about differences like drive letters and such.
The folder structure is as follows:
repo
folderToCopy
sitecore
webroot
I have tried the following using ..'s based on what tds changed my source location to be while using the "Make selected Source Location relative" option (changed it from an absolute path to ..\folderToCopy\):
../../Sitecore/Website
/../../Sitecore/Website
..\..\Sitecore\Website
\..\..\Sitecore\Website
From my understanding, TDS does the file replacement based on the files published from the associated Website project.
You can then have relative replacements such as the following:
<Replacement Include=".\assets\folderToCopy\myFile.txt">
<TargetPath>.\assets\targetFolder\myFile.txt</TargetPath>
<IsFolder>False</IsFolder>
<IsRelative>True</IsRelative>
</Replacement>
I have not been able to successfully get TDS to use the file replacement with files that are in source control but not in the project.
My suggestion would be to set up a build event that will copy these files to the correct location, or to create a nuget feed for them and pull them in as nuget references.
I am building an app for windows store and I need some default and example data to be in the localstate folder (Windows.Storage.ApplicationData.current.localFolder) when the app run the first time.
The folder and files structure is a bit complex and I tryed to copy the files at the start of the application, but I can't manage that way.
Is it possible to have files being copied automatically from the installation folder to the localstate folder during the store app installation?
Unfortunately, customization of the app install process isn't currently supported. You have to do this as part of your first run processing.
One possibility is that you include the data in your package as a .ZIP or other compressed file and use an appropriate library to expand that file into a folder structure on startup. That could simplify your logic considerably. (I don't have a library to recommend; it's just an idea.)
I copied & pasted this text here. It seems the editor seems to format some parts randomly. ;)
I downloaded ZendGdata 1.9.6, extracted it & uploaded it to my site's
root folder ..., which I need for use with Youtube API to get videos onto my site.
I must say I’m new to all this, and so I would appreciate taking this into account.
The library folder is at /ZendGdata/library.
The problem I'm having is Step. 3 when I follow instructions
(http://code.google.com/intl/de-DE/apis/gdata/articles/php_client_lib.html#gdata-installation)
for setting it up for that purpose.
Download the Google Data Client Library files.
Decompress the downloaded files. Four sub-directories should be
created:
demos — Sample applications
documentation — Documentation for the client library files
library — The actual client library source files.
tests — Unit-test files for automated testing.
Add the location of the library folder to your PHP path (see the next section)
One of the suggested locations to add the path, apart from the .htaccess file is in php.ini.
My site is on shared hosting. I have no access to the main php.ini file, but I’m allow to create one if I need one. For Drupal CMS, for some functions, it suffices placing one in the root folder.
I added this line:
include_path=".:/usr/lib/php:/usr/local/lib/php:/home/habaris6/
public_html/site.root.folder/ZendGdata/library";
When I however go to mysite.com/ZendGdata/demos/Zend/Gdata/InstallationChecker.php to test the set up, like is mentioned in the
documentation on Youtube, I get the error:
PHP Extension ErrorsTested No errors found
Zend Framework Installation Errors: Tested 0
Exception thrown trying to access Zend/Loader.php using 'use_include_path' = true.
Make sure you include Zend Framework in your include_path which currently
contains: .:/usr/lib/php:/usr/local/lib/php
SSL Capabilities Errors: Not tested
YouTube API Connectivity Errors: Not tested
So my question is: Is that the correct way to “Add the location of the library folder to your PHP path” ?
I’m a bit mixed up.
Someone was saying the php.ini file is only active in the folder where it is located. If that is the case, which of the ZendGdata folders should have it?
As I said, my purpose is to have a the Zend framework properly set up to allow using Youtube API, something I also yet have to learn to do.
In Youtube API Google group, I was referred here. The documentation coming with the downloaded file & at zend.com pre-supposes, one knows much more than some beginners like me.
Another person said I try placing this
$clientLibraryPath = '/home/habaris6/public_html/site.root.folder/ZendGdata/library';
$oldPath = set_include_path(get_include_path() . PATH_SEPARATOR . $clientLibraryPath);
in mysite.com/ZendGdata/demos/Zend/Gdata/InstallationChecker.php
Whereas everything I had tried before failed, except fot the first test, when I placed the above snippet in the installation checker, I got positive tests for everything:
Ran PHP Installation Checker on 2009-12-09T21:16:08+00:00
PHP Extension ErrorsTested: No errors found
Zend Framework Installation Errors Tested No errors found
SSL Capabilities ErrorsTested No errors found
YouTube API Connectivity ErrorsTested No errors found
Does it mean if I place that snippet in install checker, all scripts needing the library can access it?
If not, please let me know what exactly to place in the self-made php.ini & in which folder(s) it should be.
Should that not work, and I were to use .htaccess files, what exactly, based on the folders mentioned above should be the content & exactly which folders should they be in? I read that the .htaccess files should be placed in each folder. Does it really mean I should place one in each of the ZendGdata folders?
I would be grateful for any guidance enabling me to finally start, after failing to sufficient get responses elsewhere.
Thanks in advance.
It's not necessary to put all the ZendGdata code under your website document root. In fact, as a rule I don't put PHP class libraries in a location that can be accessed directly by web requests, because if there's any way to do mischief by invoking the class files directly, then anyone can do it.
Instead, put libraries outside your document root and then reference them from scripts that are run directly. For example, you could create a directory phplib as a sister to your public_html directory. Then upload the ZendGdata bundle under that phplib directory.
You can set your PHP include path in a .htaccess file. You don't need to create a .htaccess file in every directory, because the directives in any .htaccess file apply to all files and directories under the directory where the .htaccess resides. See http://httpd.apache.org/docs/2.2/howto/htaccess.html for more information.
So I would recommend creating a .htaccess file at /home/habaris6/public_html/site.root.folder containing the following directives:
<IfModule mod_php5.c>
php_value include_path ".:/usr/local/lib/php:/home/habaris6/phplib/ZendGdata/library"
</IfModule>
See http://php.net/manual/en/configuration.changes.php for more info on this.
Note that this assumes your webhosting company allows you to use .htaccess files, and that they allow you to use the php_value directive in .htaccess files. Enabling these options is an Apache configuration and they could have their own policies against that for reasons of performance or security. You should contact them for this answer; no one on the internet can answer questions about your hosting provider's policies.
If you choose to use the set_include_path() PHP function to append a directory to your runtime include path, you need to do this in each file that serves as a landing point for a web request. That is, if you permit a request to be made directly to foo.php then you need to add the code to foo.php. Any files or classes subsequently included by foo.php use the include path you defined.
Note also that whatever method you use to define the include path, it has to take effect before your script tries to load any PHP class files via the include path. The .htaccess method should accomplish this, and if you use the code method you just have to put the code high enough in your PHP script.
I don't use the method of creating a custom php.ini file under each directory within your site document tree. That's a new feature of PHP 5.3.0, not supported by earlier versions of PHP. If you're using Apache you should just use .htaccess for the same effect.