xqerl-cli

xq is a terminal cli for a dockerized xqerl)

xqerl maintained by Zachary Dean, is an Erlang XQuery 3.1 Processor and XML Database.

The small size of the dockerized xqerl make it ideal for building microservices and modern data driven websites.

This repo also a provides a simple directory structure, you can use as a template when creating xQuery projects.

Getting Started

git clone git@github.com:grantmacken/.git
cd xqerl-cli
make init
xq

make init will make sure xq is on your exec path, so you can type in xq instead of bin/xq make init also pulls down in some images. The main image is the dockerized xqerl application.

xq makes calls to a running dockerized xqerl so you need to start the container before running some commands

Note! When using make or xq stay in a xQuery project root.

make up

When up, the running containers name is xq. The terminal xq cli app is a simple executable written in bash, that communicates with the running xq container.

The xqerl database intro

The xqerl database engine can handle multiple base-URI databases. A base-URI is schema plus domain {schema}://{domain}). example http://example.com. Each base-URI constitutes a separate database.

http://example.com` # database 1
http://markup.nz`   # database 2

Each database contains collections of items referenced as URIs.

db base uri: http://example.com
db collection http://example/examples
db item http://example/examples/employees.xml

xqerl database items

The xqerl database can store

XQuery and XPath Data Model (XDM) items: These include document-nodes, arrays, maps and functions
Unparsed text items: A stored unparsed text item is one which is an item not parsed into a XDM item. Data may be extracted from unparsed text using fn:unparsed-text#1, and the data extracted via xQuery string functions and regular expressions.
link items: A db link is a reference to binary or unparsed text file on the containers file system

xqerl database

CRUD ops with xq

Create Read Update Delete

Create

xq put {file-path}
xq plonk {file-path}
xq link {file-path}

Put

xq put {file-path}

Given a path argument, the put command stores a file as a XDM item into the database, then returns the location of the stored file.

By convention all the data sources are in the src/data directory so a path can either start with a 'domain' name e.g. example.com/examples/employees.xml or use the full path src/data/example.com/examples/employees.xml. What ever the case the path must include the domain name, as the domain name becomes the base-uri part of the db stored location uri

.
└── src
    ├── data
    │   └── example.com           -> *base uri*: http://example 
    │       └── examples          -> *collection uri*: http://example/examples    
    │           ├── employees.xml -> *item uri*: http://example/examples/employees.xml

example: store employees data into xqerl database

> xq put example.com/examples/employees.xml
 - ok: stored into db
 - XDM item: document-node
 - location: http://example.com/examples/employees.xml

The xqerl database can store other XDM items beside XML documents as a document-nodes. Lets store some JSON documents.

example:

with 'mildred' JSON file store this contact data into db
with 'colors' JSON file store this list of colors data into db

> xq put src/data/example.com/examples/mildred.json
 - ok: stored into db
 - XDM item: map
 - location: http://example.com/examples/mildred.map
> xq put src/data/example.com/examples/colors.json
 - ok: stored into db
 - XDM item: array
 - location: http://example.com/examples/colors.array

As you can see from the output the 'mildred.json' doc is now a xqerl db stored 'map' item and the 'colors.json' doc is now a xqerl db stored 'map' item. Other data sources can be converted into XDM items.

*CSV *example: with entry_exit.csv store this data into db as array.

> xq put src/data/example.com/examples/entry_exit.csv
 - ok: stored into db
 - XDM item: array
 - location: http://example.com/examples/entry_exit.array

Using dockerized helpers

With some dockerized helpers, we can store other data as XDM items. Remember if we can turn our data source into db stored XDM items, then the full drill down, filter, extract and manipulate power of xPath and XQuery is available.

CommonMark to XML

note: cmark is the tool github uses to convert this README into HTML

A dockerized cmark can produce the intermediate cmark AST which is represented as a XML document. cmark uses the AST an intermediate stage to generate the HTML representation. We can stay with the produced AST XML representaion, and from this directly generate our HTML using xQuery.

**** example: with index.md convert into cmark XML then store into db as a document-node.

> xq put src/data/example.com/content/index.md
 - ok: stored into db
 - XDM item: document-node
 - location: http://example.com/content/index.cmark.xml

Note: TODO! - another section on cmark XML to HTML conversion using xQuery modules

html-tidy HTML to XML

A dockerized htmltidy can produce from a HTML source, a well formed XML document.

**** example: store hello-world.html into db as a document-node.

> xq put src/data/example.com/examples/hello-world.html
 - ok: stored into db
 - XDM item: document-node
 - location: http://example.com/examples/hello-world.xhtml

Plonk

Command: xq plonk {text-file}

Given a path argument, the plonk command stores a text file into the database, then returns the location of the stored unparsed text item.

The raw text is just plonked into the database bypassing the 'file to XDM item conversion' as done by xq put. The raw text will be stored as the xs:string XDM data type and the resource can be retrieved using the xQuery fn:unparsed-text#1 function.

Data may be extracted from a string, using xQuery string functions as well as xQuery matches#2,replace#3 and tokenize#2 regular expressions

Use cases: You might want to store a markdown document 'as is' so it can be inserted into a HTML textarea.

Example: store a markdown document as a xs:string XDM item

> xq plonk example.com/content/index.md
http://example.com/content/index.md
 - ok: stored into db
 - XDM item:  xs:string
 - location: http://example.com/content/index.md

Link

Command: xq link {domain} {asset-path}

Given a 'domain' and 'path' argument, the xq link command creates a link in the xqerl database to a preprocessed asset located on the containers file system.

TODO: link to why this is a good thing.

All asset sources are located in the ./src/static_assets/ directory so the {asset-path} will be resolved as ./src/static_assets/icons/article.svg

Before the asset is stored the file can be pipelined thru docker container instances to get a preferred outcome. For static assets this outcome usually means some form file size reduction.

By convention all the data sources are in the src/data directory, and it is no surprise that static asset sources are located in the ./src/static-assets/ directory

Preprocessing pipeline example which produces a gzipped svg file with a svgz extension

article.svg =>
scour =>
zopfli =>
article.svgz

The xq link command will produce two outcomes.

a binary asset on the static-assets container volume. The static-assets container volume is mounted on the xqerl priv/static container directory. priv/static/icons/article.svgz
A db link to the file asset. example: the db link 'http://example.com/icons/article' points to the file 'file:///usr/local/xqerl/priv/static/icons/article.svgz'

note: links are searchable db items

example: create db link to compressed svgz file

> xq link example.com icons/article.svg
1. file: /usr/local/xqerl/priv/static/assets/icons/article.svgz
2. db link: http://example.com/icons/article.svgz

example: create db link to gzipped js file

> xq link example.com scripts/prism.js
1. file: /usr/local/xqerl/priv/static/assets/scripts/prism.js.gz
2. db link: http://example.com/scripts/prism.js.gz

example: create db link to gzipped css file

> xq link example.com src/static_assets/styles/index.css
1. file: /usr/local/xqerl/priv/static/assets/styles/index.css.gz
2. db link: http://example.com/styles/index.css.gz

example: create db link to font woff file

> xq link example.com fonts/ibm-plex-mono-v5-latin-regular.woff2
1. file: /usr/local/xqerl/priv/static/assets/fonts/ibm-plex-mono-v5-latin-regular.woff2
2. db link: http://example.com/fonts/ibm-plex-mono-v5-latin-regular.woff2

Read or Retrieve

xq list {db-uri}
xq available {db-uri}
xq get {db-uri}
xq get {db-uri} {xpath-or-lookup}
xq get {db-uri} {xpath-or-lookup} {bang-or-arrow}
WIP: xq collect {db-uri}

List

Command: xq list {db-uri}

example: lists items available in 'examples' collection

> xq list example.com/examples
http://example.com/examples/employees.xml
http://example.com/examples/hello-world.xhtml
http://example.com/examples/works.xml
http://example.com/examples/colors.array
http://example.com/examples/entry_exit.array
http://example.com/examples/mildred.map

Available

Command xq available {db-uri}

example: is colors.array available in db

> xq available example.com/examples/colors.array
true

Get Item

xq get {db-uri} where the arg {db-uri} is an item in the database The command returns a serialized representation of a item. For document-nodes this will be a XML string. For arrays or maps this will be a JSON string.

example: get a document-node

> xq get example.com/examples/employees.xml
<employees>
    <employee>
        <employeeId>4</employeeId>
        <reportsTo>1</reportsTo>
        <name>Charles Madigen</name>
        <job>Chief Operating Officer</job>
        <Phone>x10962</Phone>
        <email>cmadigan@example.com</email>
        <department>Management</department>
        <salary>26200.00</salary>
        <gender>male</gender>
        <maritalStatus>married</maritalStatus>
        <employeeType>full time</employeeType>
    </employee>
</employees>

example: get an array item

> xq get example.com/examples/colors.array | jq '.'
[
  {
    "color": "Green"
  },
  {
    "color": "Pink"
  },
  {
    "color": "Lilac"
  },
  {
    "color": "Turquoise"
  },
  {
    "color": "Peach"
  },
  {
    "color": "Opal"
  },
  {
    "color": "Champagne"
  }
]

Get Item Then

With db uri, get XDM item then depending on item use xPath or lookup expression to drill down or filter, then optionally use 'arrow' or 'bang' to modify, to return serialized item or items

document node commands

xq get {db-uri} {xpath}

Get document node, then apply xpath expression, to return a serialized XML string.

example: extract first employees name

> xq get example.com/examples/employees.xml \
> '//employee[1]/name/string()'
Charles Madigen

Command: xq get {db-uri} {xpath} {bang}

Get document node, then apply xpath expression, then use bang (simple map expression) to return result.

example: list active employees

> xq get example.com/examples/works.xml \
> '//employee[./status ="active"]' \
> '! concat(./@name/string(), " - " , ./status/string())'
Jane Doe 13 - active

Command: xq get {db-uri} {xpath} {arrow}

Get document node, then get nodes with xpath expression, then use arrow expression to return result.

example: total hours worked

> xq get example.com/examples/works.xml '//employee/hours' '=> sum() => string()'
592

array and map commands

Command: xq get {db-uri} {lookup}

Get map or array, then use lookup expression to return object as a serialized JSON string.

examples with the Mildred map drill down to get address and town

> xq get example.com/examples/mildred.map '?address'
{"county":"Oxfordshire","postcode":"OX6 3PD","street":"91 High Street","town":"Biscester"}
> xq get example.com/examples/mildred.map '?address?town'
"Biscester"

Command: xq get {db-uri} {lookup_expr} {bang_expr}

Get map or array, then get object with lookup, then use bang expression to return a serialized JSON string.

example: list colors without Lilac

> xq get example.com/examples/colors.array '?*' '! ( .?color[not(. = "Lilac")] )'
["Champagne",
 "Green",
 "Opal",
 "Peach",
 "Pink",
 "Turquoise"]

Command: xq get {db-uri} {bang-expr}

Get map or array, then use bang expression to return a serialized JSON string.

example: format mildreds' name

> xq get example.com/examples/mildred.map '! ``[ `{.?firstname}` `{.?lastname}`]``'
 Mildred Moore

Update

' xq update {db-uri} [insert, replace, delete, rename ]'

xqerl implements the xQuery update facility

insert
replace
delete
rename

Update Insert

example: add more hours for employee[3]

xq get example.com/examples/works.xml '//employee[3]/hours => sum()'
xq update example.com/examples/works.xml insert node '<hours>40</hours>' into '//employee[3]'
xq get example.com/examples/works.xml '//employee[3]/hours => sum()'

Update Delete

Update db document, with delete expression

xq update {db-uri} delete ...

example delete first lot of hours for employee[2]

xq get example.com/examples/works.xml '//employee[2]/hours => sum()'
xq update example.com/examples/works.xml delete node '//employee[2]/hours[1]'
xq get example.com/examples/works.xml '//employee[2]/hours => sum()'

Update Replace

Update db document, with replace expression

xq update {db-uri} replace ...

example correct the hours for employee[1]

xq get example.com/examples/works.xml '//employee[1]/hours => sum()'
 xq update example.com/examples/works.xml replace node '//employee[1]/hours[1]' with '<hours>25</hours>'
xq get example.com/examples/works.xml '//employee[1]/hours => sum()'

Update Rename

Update db document, with rename expression

xq update {db-uri} rename ...

example rename first employee tag to contractor

xq update example.com/examples/works.xml rename node '/works/*[1]' as '"contactor"'
xq get example.com/examples/works.xml '//*[1]

Delete

Command: xq delete {db-uri} deletes item in collection

example delete colors array

> xq delete example.com/examples/colors.array
 - deleted array item [ http://example.com/examples/colors.array ]

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

README.md

README.md

xqerl-cli

Getting Started

The xqerl database intro

xqerl database items

xqerl database

Create

Put

Using dockerized helpers

CommonMark to XML

html-tidy HTML to XML

Plonk

Link

Read or Retrieve

List

Available

Get Item

Get Item Then

document node commands

array and map commands

Update

Update Insert

Update Delete

Update Replace

Update Rename

Delete

Files

README.md

Latest commit

History

README.md

File metadata and controls

xqerl-cli

Getting Started

The xqerl database intro

xqerl database items

xqerl database

Create

Put

Using dockerized helpers

CommonMark to XML

html-tidy HTML to XML

Plonk

Link

Read or Retrieve

List

Available

Get Item

Get Item Then

document node commands

array and map commands

Update

Update Insert

Update Delete

Update Replace

Update Rename

Delete