Overview of HTSQL¶
Table of Contents
HTSQL was created in 2005 to provide an XPath-like HTTP interface to PostgreSQL for client-side XSLT screens and reports. HTSQL found its audience when analysts and researchers bypassed the user interface and started to use URLs directly. The language has evolved since then.
What is HTSQL?¶
HTSQL is a comprehensive navigational query language for relational databases and web service gateway.
HTSQL is a Web Service¶
http://demo.htsql.org/school
On the left is a URL, on the right is what a browser would show.
HTSQL is a query language for the web. Queries are URLs that can be directly typed into a browser; the output could be returned in a variety of formats including HTML, CSV, JSON, etc. HTSQL can be used as the basis for dashboarding tools and other browser-based applications. In this way, database queries can be shared, tweaked, and used in any number of ways.
HTSQL is a Relational Database Gateway¶
SELECT "school"."code",
"school"."name",
"school"."campus"
FROM "ad"."school"
ORDER BY 1 ASC
On the left is an HTSQL query, on the right is SQL it is translated to.
HTSQL wraps your existing existing relational database, transparently handling SQL complexities for you. The current version of HTSQL supports SQLite, PostgreSQL, MySQL, Oracle, and Microsoft SQL Server. We’ve taken care to abstract differences between these SQL dialects so that a given HTSQL query has consistent semantics across database server implementations.
HTSQL is an Advanced Query Language¶
SELECT "school"."name", COALESCE("program"."count", 0), COALESCE("department"."count", 0)
FROM "ad"."school"
LEFT OUTER JOIN (SELECT COUNT(TRUE) AS "count", "program"."school_code" FROM "ad"."program" GROUP BY 2) AS "program" ON ("school"."code" = "program"."school_code")
LEFT OUTER JOIN (SELECT COUNT(TRUE) AS "count", "department"."school_code" FROM "ad"."department" GROUP BY 2) AS "department" ON ("school"."code" = "department"."school_code")
ORDER BY "school"."code" ASC
On the left is an HTSQL query, on the right is SQL it is translated to.
Besides typical expression algebra and function set, HTSQL provides sophisicated navigational query mechanism, composable query fragments and an extensive macro inclusion system. In particular, nested aggregations and projections are easy to understand and use.
HTSQL is a Communication Tool¶
Show me schools, and, for each school,
- its name, its location,
- number of programs and departments,
- and the average number of courses
across each of its departments?
On the left is a business inquiry, on the right is the HTSQL translation.
HTSQL is first and formost designed for the accidental programmer and as such provides a direct mapping of common business inquiries onto a computer parsable and executable syntax. Just because a query must be processable by a machine, doesn’t mean it shouldn’t be human readable.
HTSQL is a Python Library¶
from htsql import HTSQL
conn = HTSQL("pgsql:///htsql_demo")
rows = conn.produce("/school")
for row in rows:
print row
school(code=u'art',
name=u'School of Art & Design',
campus=u'old')
school(code=u'bus',
name=u'School of Business',
campus=u'south')
...
HTSQL can be embedded into any Python application to provide an intuitive object based query engine for complex reporting. It works out of the box with Jinja and other tools. We provide meta-data adapters for Django and SQLAlchemy.
Our Philosophy¶
We develop HTSQL to liberate the Accidental Programmer—professionals and data experts who are not software engineers by trade, but who must write database queries or data processing code to get things done. HTSQL handles routine data processing needs in an accessible, transparent, rigorous and embeddable manner.
Accessible¶
We want HTSQL to be broadly usable. Our query language should provide business analysts, information scientists, and data curators self-service access and control over their database. Not only should these accidental programmers be able able to answer complex business inquiries themselves, they should be able to share the queries they create with their colleagues. HTSQL must be a productive tool for problem solving with a small learning curve.
Transparent¶
We think aesthetics matter. When a data analyst is focusing on a domain specific problem, the HTSQL query language should do the heavy lifting but otherwise stay in the background. Translation of a business inquiry into the HTSQL query language must be natural and obvious. Initial query authoring is just the beginning. A database query is often the only human readable expression of a business rule, so each and every query must be a pleasure to review, share and maintain.
Rigorous¶
We know correctness is critical. The HTSQL query language is based upon a navigational data linking and flow processing model having consistent semantics that are independent of the underlying database architecture. The language designed to be composable so that query fragments can be independently tested and combined. HTSQL’s syntax is regular enough that syntax highlighting and context sensitive name lookup is possible.
Embeddable¶
We realize a query language is not a product. Instead, HTSQL is a tool used as part of a workflow solution or embedded into an application. Our public application program interface is be simple and stable. Further, since accidental programmers might require features which other software developers could build, we have an extensive plug-in interface for those who wish to add features to HTSQL itself. Between web service wrappers or plug-ins, it should be possible to customize almost every aspect of HTSQL without requiring a code fork.
HTSQL in a Nutshell¶
HTSQL was designed from the ground up as a self-serve reporting tool for data analysts. With HTSQL, the easy stuff is truly easy; and, the complex stuff is easy too.
In this section we introduce the fundamentals of HTSQL syntax and semantics. For a more incremental approach, please read the HTSQL Tutorial. For the purposes of this section, we use a fictitious university schema.
This data model has two top-level tables, school and department, where department has an optional link to school. Subordinate tables, course and program, have mandatory links to their parents.
Scalar Expressions¶
Literal values:
| 3.14159 | ‘Hello World!’ |
|---|---|
| 3.14159 | Hello World! |
Algebraic expressions:
Predicate expressions:
| (7<13)&(1=0|1!=0) |
|---|
| true |
Filtering¶
Sieve operator produces records satisfying the specified condition:
| code | name | campus |
|---|---|---|
| bus | School of Business | south |
| mus | School of Music & Dance | south |
Sorting operator reorders records:
| code | name | campus |
|---|---|---|
| ph | Public Honorariums | — |
| sc | School of Continuing Studies | — |
| eng | School of Engineering | north |
| art | School of Art and Design | old |
| … | … | … |
Truncating operator takes a slice from the record sequence:
| code | name | campus |
|---|---|---|
| art | School of Art and Design | old |
| bus | School of Business | south |
Selection & Definition¶
Selection specifies output columns:
| name | campus |
|---|---|
| School of Art and Design | old |
| School of Business | south |
| College of Education | old |
| School of Engineering | north |
| … | … |
Title decorator defines the title of an output column:
| name | # of Dept |
|---|---|
| School of Art and Design | 2 |
| School of Business | 3 |
| College of Education | 2 |
| School of Engineering | 4 |
| … | … |
Calculated attributes factor out repeating expressions:
| code | num_dept |
|---|---|
| eng | 4 |
| la | 5 |
| mus | 4 |
| ns | 4 |
References carry over values across nested scopes:
| title | credits |
|---|---|
| Financial Accounting | 5 |
| Audit | 5 |
| Accounting Internship | 6 |
| History of Art Criticism I | 4 |
| … | … |
Aggregation¶
Aggregates convert plural expressions to singular values.
Scalar aggregates:
| count(department) |
|---|
| 27 |
Nested aggregates:
| avg(school.count(department)) |
|---|
| 2.66666666667 |
Various aggregation operations:
| name | count(course) | max(course.credits) | sum(course.credits) | avg(course.credits) |
|---|---|---|---|---|
| Accounting | 12 | 6 | 42 | 3.5 |
| Art History | 20 | 6 | 70 | 3.5 |
| Astronomy | 22 | 5 | 66 | 3.0 |
| Bioengineering | 17 | 8 | 55 | 3.23529411765 |
| … | … | … | … | … |
Projection¶
Projection (^) returns distinct values. This example returns distinct campus values from the school table:
| campus |
|---|
| north |
| old |
| south |
In the scope of the projection, school refers to all records from school table having the same value of campus attribute:
| campus | count(school) |
|---|---|
| north | 1 |
| old | 4 |
| south | 2 |
Projections combine with other language features in a natural way. The next example displays the number of distinct program degrees offered by each school:
| name | count(program^degree) |
|---|---|
| School of Art and Design | 2 |
| School of Business | 4 |
| College of Education | 4 |
| School of Engineering | 2 |
| … | … |
Linking¶
Links between tables are automatic and relative, inferred from foreign key constraints. Unlink (@) permits arbitrary, non-relative linking.
| name | count(department) |
|---|---|
| School of Business | 3 |
| School of Engineering | 4 |
| School of Arts and Humanities | 5 |
| School of Music & Dance | 4 |
| … | … |
The query above returns schools with the number of departments above average among all schools.
Why not SQL?¶
Relational algebra is frequently inadequate for encoding business inquiries — elementary set operations do not correspond to meaningful data transformations. The SQL language itself is tedious, verbose, and provides poor means of abstraction. Yet, the relational database is an excellent tool for data modeling, storage and retrieval.
HTSQL reimagines what it means to query a database. The combination of a navigational model with data flows enables expressions that naturally reflect business inquiries. The HTSQL translator uses SQL as a target assembly language, which allows us to fix the query model and language while keeping current investment in relational systems.
SQL Conflates Rows & Columns¶
“For each department, please show the department name and the corresponding school’s campus.”
This business inquiry clearly separates the requested rows (each department) and columns (department name and corresponding school’s campus), but this separation is lost when the query is encoded in SQL:
SELECT d.name, s.campus
FROM ad.department AS d
LEFT JOIN ad.school AS s
ON (d.school_code = s.code);
In this SQL query, the FROM clause not only picks target rows, but also includes extra tables required to produce output columns. This conflation makes it difficult to determine business entities represented by each row of the output.
The HTSQL translation separates the row definition from the column selection. The linking is implicit, and correct. The encoded query can be read aloud as a verbal inquiry.
SQL Conflates Filters & Links¶
“For each department, return the department’s name and number of courses having more than 2 credit hours.”
This business inquiry returns department records, and for each record summarizes associated courses meeting a particular criteria.
SELECT d.name, COUNT(SELECT TRUE FROM ad.course AS c
WHERE c.department_code = d.code
AND c.credits > 2)
FROM ad.department AS d;
For this SQL encoding, the WHERE clause of the subquery conflates the linking of course to department with the filter criteria.
SELECT d.name, COUNT(c)
FROM ad.department AS d
LEFT JOIN ad.course AS c
ON (c.department_code = d.code
AND c.credits > 2)
GROUP BY d.name;
In a common optimization, the correlated subquery is replaced with a GROUP BY projection. This encoding further obfuscates the business inquiry by conflating in two ways — row/column and link/filter.
The HTSQL translation keeps the filter criteria separate from linking and the row definition separate from output columns. The query adheres the form of the original business inquiry.
Conflating Projection with Aggregation¶
“How many departments by campus?”
This business inquiry asks for rows corresponding to each campus, and for each row, the number of correlated departments. In the schema, there isn’t a campus table, so we have to take distinct values of campus column from the school table. This operation is called projection.
SELECT s.campus, COUNT(d)
FROM ad.school AS s
LEFT JOIN ad.department AS d
ON (s.code = d.school_code)
WHERE s.campus IS NOT NULL
GROUP BY s.campus;
For this SQL encoding, the GROUP BY clause combines two operations: projection and evaluating the aggregate COUNT(). This conflation causes a reader of the query some effort determining what sort of rows are returned and how the aggregate is related to those rows.
In the HTSQL query, we start with an explicit projection (the ^ operator), then we select correlated columns. This way, the aggregation is indicated separately as part of the column selector rather than being conflated with the row definition.
SQL Lacks Means of Encapsulation¶
“For each department, return the department name and the number of offered 100’s, 200’s, 300’s and 400’s courses.”
In this business inquiry, we are asked to evaluate the same statistic across multiple ranges.
SELECT d.name,
COUNT(CASE WHEN c.no BETWEEN 100 AND 199 THEN TRUE END),
COUNT(CASE WHEN c.no BETWEEN 200 AND 299 THEN TRUE END),
COUNT(CASE WHEN c.no BETWEEN 300 AND 399 THEN TRUE END),
COUNT(CASE WHEN c.no BETWEEN 400 AND 499 THEN TRUE END)
FROM ad.department AS d
LEFT JOIN ad.course AS c
ON (c.department_code = d.code)
GROUP BY d.name;
This query is tedious to write and error prone to maintain since SQL provides no way to factor the repetitive expression COUNT(...).
The HTSQL translation avoids this duplication by defining a calculated attribute count_courses($level) on the department table and then evaluating it for each course level.
In SQL, Modest Complexity is Painful¶
“For each school with a degree program, return the school’s name, and the average number of high-credit (>3) courses its departments have.”
This business inquiry asks us to do the following:
- pick records from the school table
- keep only those with an associated degree program
- for each school record, compute average of:
- for each associated department, count:
- associated courses with credits>3
- for each associated department, count:
SELECT s.name, o.avg_over_3
FROM ad.school AS s
JOIN ad.program AS p ON (p.school_code = s.code)
LEFT JOIN (
SELECT d.school_code, AVG(COALESCE(i.over_3,0)) AS avg_over_3
FROM ad.department d
LEFT JOIN (
SELECT c.department_code, COUNT(c) AS over_3
FROM ad.course AS c WHERE c.credits > 3
GROUP BY c.department_code
) AS i ON (i.department_code = d.code)
GROUP BY d.school_code
) AS o ON (o.school_code = s.code)
GROUP BY s.name, o.avg_over_3;
Not only is this SQL encoding is hard to read, it took several passes to get right — without the COALESCE you get results that look correct, but aren’t.
Each syntactic component of the HTSQL query is self-contained; when assembled, they form a cohesive translation of the business inquiry.