buda-base / bdrc-lib-app Goto Github PK

View Code? Open in Web Editor NEW

5.0 5.0 2.0 42.76 MB

JavaScript 98.13% CSS 1.87%

bdrc-lib-app's Introduction

Vagrant scripts for BUDA platform instanciation

The base platform is built using Vagrant and VirtualBox:

Install Vagrant and VirtualBox.
Download or git clone this repository.
cd into the unzipped directory or git clone
install VirtualBox guest additions with vagrant plugin install vagrant-vbguest
run vagrant up to summon a local instance

Or for an AWS EC2 instance:

install the vbguest plugin: vagrant plugin install vagrant-vbguest
and run the command: vagrant up or rename Vagrantfile.aws to Vagrantfile and run vagrant up --provider=aws

This will grind awhile installing all the dependencies of the BUDA platform.

Once the initial install has completed the command: vagrant ssh will connect to the instance where development, customization of the environment and so on can be performed as for any headless server.

Similarly, the jena-fuseki server will be listening on:

http://localhost:13180/fuseki

Lds-pdi application is accessible at :

http://localhost:13280/

(see https://github.com/buda-base/lds-pdi/blob/master/README.md for details about using this rest services)

The command: vagrant halt will shut the instance down. After halting (or suspending the instance) a further: vagrant up will simply boot the instance without further downloads, and vagrant destroy will completely remove the instance.

If running an AWS instance, after provisioning access the instance via ssh -p 15345 and delete Port 22 from /etc/ssh/sshd_config and sudo systemctl restart sshd. This will further secure the instance from attacks on port 22.

bdrc-lib-app's People

Contributors

Stargazers

Watchers

Forkers

sonam152 danzinxirao

bdrc-lib-app's Issues

Background image in Google Play

Recent version of the image with stacked letters displaying properly should be uploaded. Here's the current version:

Emptying the app cache fixes it ... 1 time.

App is frozen as described in #38:

App cache is emptied:

I'm pretty sure the data comes from images loaded for reading

Query 1 and 2 work fine:

Query 3 crashes the app:

Tried to clear the cache again but it's empty and doesn't make a difference:

Rename app to "BDRC Mobile"

"དཔེ་ཀློག་མཉེན་ཆས། BDRC Lib" or "BDRC Lib དཔེ་ཀློག་མཉེན་ཆས།" to make it more discoverable for the 6 million Tibetans

Publication Information is needed

NT: It’s crucial for identification, title and authorship aren’t enough in many cases.
Karma: Need full publication information show.

Crash on index import

It appears that larger index files cause a memory crash on some phones. This is the same behaviour that was initially seen with very large index files. The suggestion is to further break the files down into 1MB or 2MB each. What do you think @eroux ?

Search by type

NT: Search icons are just a random suggestion. I’m not sure filtered search is the right way to go, organizing the search results as on tbrc.org might be better. I’m sure Chris has thought this through and knows what is the better strategy. Starting with an empty page is quite nice too...

Can't find the app on App Store

I've tried on the Taiwanese App Store as well as the Indian one. Or is it that it's not there yet?

NT: “At least 2 syllables” is useful info for first time users, but might not need to be in the way afterwards.

Also suggesting replacing the Tibetan "Search" hint in the search box with the 2 syllable requirement line.

Work Index Files Contain Publisher Names

NT discovered that the search results contained publisher names as well as work titles and alternative work titles (or outline titles or author names). Looking at the index file, it appeared that all of these results were connected to values containing the work id twice... like this:

  "གསང་ཆེན་བསྟན་པའི་ཆུ་འགོ་རྒྱལ་བ་ཀཿ་ཐོག་པའི་ལོ་རྒྱུས་མདོར་བསྡུས་བརྗོད་པ་འཆི་མེད་ལྷའི་རྔ་སྒྲ་ངོ་མཚར་རྣ་བའི་དགའ་སྟོན": [
    "W20396-W20396"
  ]

(NOTE: I'm not sure whether this example is a publisher name or alternative title, it is just for illustration)

The result was that there was no detail file for any of the alternate titles (W20396-W20396.json did not exist, but W20396.json did, of course), or for publisher names. I added code that just used the first ID, which took care of the missing data problem. However, it still did not look good to @ngawangtrinley to have publisher names with icons for works. Given that we could not figure out how to differentiate, NT determined it was best to drop all of the alternative titles and publisher names (anything with a dash in the RID). So that is what the code does.

App Base Requirements

Lots of space, IOS 9+,

Karma: App fails on initialization . He only has 282 MB of space. Now, App is working fine. I deleted some images and video clips from my iphone to make space.

Single Creator name in Work View

Asa: Use the first author in the list.

As shown in the above example, variants of an author's names are displayed in a work view (W1KG25095), and make users strain their thumbs scrolling down the list.

This name:
<work:creator person="P270">o rgyan 'jigs med chos kyi dbang po</work:creator>
should be displayed in a Work View.

The variants can appear in the Person View.
{"creatorOf":["W8LS19616","W8LS19323","W2CZ5981","W30038","W1KG24057","W1KG25095","W4PD913","W5741","W8LS17113","W00KG04062","W5743","W8LS24641","W8LS20832","W8LS19604","W21707","W8LS19422","W3CN9771","W5832","W24829","W1PD107142","W1KG21887","W1KG24014","W1KG2820","W8LS19972","W22270","W1KG3552","W21999","W1KG3818","W8LS16587","W3CN5420","W8LS16483","W1KG23850","W2CZ6659","W1KG12208","W1KG13014","W1KG3401","W3CN5444","W3PD441","W8LS19437","W8LS17020","W23370","W1PD45158","W4CZ299952","W21857","W8LS18308","W3CN4690","W1KG1578"],"name":["རྫོགས་ཆེན་མཁན་རབས ༠༥","ཨོ་རྒྱན་འཇིགས་མེད་ཆོས་ཀྱི་དབང་པོ།","ཨོ་རྒྱན་འཇིགས་མེད་ཆོས་དབང་།","རྫ་དཔལ་སྤྲུལ་ཨོ་རྒྱན་འཇིགས་མེད་ཆོས་ཀྱི་དབང་པོ།","རྫོགས་ཆེན་མཁན་རབས ༠༥ ཨོ་རྒྱན་འཇིགས་མེད་ཆོས་ཀྱི་དབང་པོ།","ཁྱི་རྒན་དཔལ་དགེ"]}

Tibetan Typos

There are some typos: སྙེ་ཆས། need to be corrected to མཉེན་ཆས།

Tibetan word ཡིག་འབྲུ་གཉིས་ means letter and here need syllable instead of letter, therefore, correct word will be either of ཚེག་ཁྱིམ་གཉིས་ ཚེག་བར་གཉིས་

Redeploy with Data 2017.07.02

Download and include the latest dataset: data-20170712-2.zip

Person is found but is empty

Query:

Hits:

Info after selecting the person:

Customer service wechat in info

Add a QR code or link to a customer service account in info. The account can be followed by Dondrup during the day in Asia, and by Emma during the American day

Confusing message during initial search

Users get confused when they read "Search results: 0" while the spinner still spins.

User Guide

Karma/Librarians: User guideline would be very helpful to the new user of the BDRC Lib App.
Such as explanation of the icon to identify them.

Optimize Search by recognizing subsets of previous search

When the first "valid" search has been performed, a subset of the database is returned, which can now simply be reduced, instead of making queries to the database. To make this very fast it would be good to keep each subset. This works well in english, but need to investigate how it works in Tibetan with stacked glyphs.

label changes

As seen in the Skype discussion, @karmagongde requested to change the labels in the following way:

བརྩམས་ཆོས་ངོ་རྟགས into མི་སྣའི་ཨང་རྟགས
མཛད་པ་པོ into བརྩམས་ཆོས་ཁག

@ngawangtrinley can you please confirm?

Show Primary Author Only

Karma/Librarians: When the author name is displayed, only the primary name would be enough. Rest could be hide/show.

Clean the cache if App stopped working.

Karma/Librarians: Clean the cache if App stopped working.

More info is needed on this.

Additional Navigation For Viewing Texts/Works

Karma/Librarians: For the images, if there are four buttons same as in our website would be useful to go end page, check couple of pages backward, Then able to Back to the Beginning page.

Android BDRC Launch Screen Missing

The launch screen for Android needs to be configured and added to the Android deployed project.

App screenshots

The publisher name on the screenshots of the app (both on Play and AppStore) is some English text displayed in Tibetan characters. Covfefe-esque.

To be more precise, it is the Tibetan font rendering of an English transliteration of the Tibetan name of the publisher. I know I should have noticed this earlier.

Minimum Publication Information

Publisher Name
Publisher Location
Publish Date

Search Results, Search icon reminds user what is being searched

separating data?

I don't think we can keep the data in the git repository in the long term (although we can for release 1), so maybe I could just upload the data on a public s3 url and you could just reference it in the installation instructions?

Copyright notice

Some notification when you are not online (and go to view images)

Detail View: Replace RID header with title

NT: Replace the RID in header by title, for non-librarian terrestrials

Volumes order

Volumes listed for a work aren't ordered.

Alternative titles of the same work appear as different hits

Can get a bit confusing for users at first. A visual hint showing that they're the same would help.

Dirk: App icon design is too square

Small thing: App icon design is too square -- all other icons have rounded corners, so it looks off

Suggest to the users, searching with more words will yield more precise results.

Karma/Librarians

Status reflected in search results

NT: This one is quite important imo. Icons with type of result and status. Proposed set of icons below...

Identifier and fields are all disorder.

Karma/Librarians: Identifier and fields are all disorder. See the screenshot as example.

Search Spinner

Add some sort of indication that results are still being loaded.

If the full title exists in BDRC website/App

Karma/Librarians: The good chance is the text is available in BDRC.

Detail View: NT Interface and data display ideas

Cover page for quick identification, 80% of the time a glance at the cover page will be enough to identify the work and edition. Pecha layout needed.

Headers/content mix-up

Outline view to access texts (RID in banner should be work title, and RID should be displayed with Publication Information)

Comments improve the communication between ground staff and librarians, and users and librarians. This one is very very important. :D

Another random suggestion. Everything as above but publication information is accessed through the info icon, this brings down the amount of scrolling needed to reach the bottom of the page. The same thing can be done by collapsing data in each section though.

Missing Images

Images in legacy image groups that required normalization are causing an exception in the image service and not displaying in the App. Example:

https://www.tbrc.org/browser/ImageService?work=W22084&igroup=I0981&image=1&first=1&last=530&fetchimg=yes

The image normalization code was removed from the app when the ImageService was updated to handle the normalization.

U+2028 (LINE SEPARATOR) in index files appears to cause a fatal database error on resolution of a query

This issue is likely related to this issue:

storesafe/cordova-sqlite-storage#147

It appears that whenever a query returns results that contain one of the many indices where the key has the character combination '\n' -- a crash happens. Here is an example.

"གྲགས་པ་རྒྱལ\nམཚན། གསུང་འབུམ":["W3GSP297"]

The temporary fix is to remove all of these from the index files.

Noto Sans for English Toolbar titles (Not Noto Sans Tibetan)

This will work better than Noto for English, centering properly.

swapped labels

publisherName should be པར་གནས་ and publisherDate should be པར་དུས་, they are swapped now

App crashes after a couple of queries

Query A: "བདེ་ལེགས་"

Query A results:

Query B: "དོན་གྲུབ་རྒྱལ་"

Repeat of query A:

Repeat of query A after restarting my phone:

Query C "དོན་གྲུབ་" before all results of previous query could be displayed:

Query A again:

And query C after restarting:

Quickly followed by query D "ཨེ་མ་ཧོ":

easing testing

I tried to test the app in a browser to check my improvements of #4. I think it could be a bit easier: after I ran yarn and npm run dev, I got

ERROR in [copy-webpack-plugin] unable to locate '/home/eroux/BUDA/softs/BDRC-Lib-App/www_src/onsenui_css_theming_tools/build/css' at '/home/eroux/BUDA/softs/BDRC-Lib-App/www_src/onsenui_css_theming_tools/build/css'
Child extract-text-webpack-plugin:
       [0] ./~/css-loader/lib/css-base.js 1.46 kB {0} [built]
       [1] ./~/css-loader?importLoaders=1!./~/postcss-loader!./src/app/DetailPage.pcss 407 bytes {0} [built]
Child extract-text-webpack-plugin:
       [0] ./~/css-loader/lib/css-base.js 1.46 kB {0} [built]
       [1] ./~/css-loader?importLoaders=1!./~/postcss-loader!./src/app/Top.pcss 871 bytes {0} [built]
Child extract-text-webpack-plugin:
       [0] ./~/css-loader/lib/css-base.js 1.46 kB {0} [built]
       [1] ./~/css-loader?importLoaders=1!./~/postcss-loader!./src/app/SearchPage.pcss 209 bytes {0} [built]
Child extract-text-webpack-plugin:
       [0] ./~/css-loader/lib/css-base.js 1.46 kB {0} [built]
       [1] ./~/css-loader?importLoaders=1!./~/postcss-loader!./src/app/SearchResults.pcss 484 bytes {0} [built]

and then nothing really happens: the process continues to run without printing anything, but there seems to be no open port waiting for me to test the app... what can I do?

Open images on page 3

Open a book on cover page, the 2 BDRC pages will still be there but users won't have to wait for them to load to get to the the cover page. It can make a big difference with a slow connection.

sorting results

Results should be sorted (see google doc). A solution that seems reasonable is to use Intl.Collator with the bo locale, or maybe falling back to the dz locale for very old phones (the bo locale should work for 2 years or so).

As a side not there is a request to improve tibetan collation in CLDR with the work I did in this repo.

I'm running some experiments on http://eroux.fr/locales.html it doesn't seem to work, but I cannot understand why, i'll file some bugs, ask questions, etc. and update the issue. In the meantime I'm interested by the result of the page in your browser and on your phones!