Adding upstream version 10.5.2.upstream/10.5.2

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 3 insertions, 3 deletions

diff --git a/posts/python_sql_engine.md b/posts/python_sql_engine.md
index 1c74680..01ad71d 100644
--- a/posts/python_sql_engine.md
+++ b/posts/python_sql_engine.md
@@ -14,13 +14,13 @@ This post will cover [why](#why) I went through the effort of creating a Python
 * [Executing](#executing)
 
 ## Why?
-I started working on SQLGlot because of my work on the [experimentation and metrics platform](https://netflixtechblog.com/reimagining-experimentation-analysis-at-netflix-71356393af21) at Netflix, where I built tools that allowed data scientists to define and compute SQL-based metrics. Netflix relied on multiple engines to query data (Spark, Presto, and Druid), so my team built the metrics platform around [PyPika](https://github.com/kayak/pypika), a Python SQL query builder. This way, definitions could be reused across multiple engines. However, it became quickly apparent that writing python code to programatically generate SQL was challenging for data scientists, especially those with academic backgrounds, since they were mostly familiar with R and SQL. At the time, the only Python SQL parser was [sqlparse]([https://github.com/andialbrecht/sqlparse), which is not actually a parser but a tokenizer, so having users write raw SQL into the platform wasn't really an option. Some time later, I randomly stumbled across [Crafting Interpreters](https://craftinginterpreters.com/) and realized that I could use it as a guide towards creating my own SQL parser/transpiler.
+I started working on SQLGlot because of my work on the [experimentation and metrics platform](https://netflixtechblog.com/reimagining-experimentation-analysis-at-netflix-71356393af21) at Netflix, where I built tools that allowed data scientists to define and compute SQL-based metrics. Netflix relied on multiple engines to query data (Spark, Presto, and Druid), so my team built the metrics platform around [PyPika](https://github.com/kayak/pypika), a Python SQL query builder. This way, definitions could be reused across multiple engines. However, it became quickly apparent that writing python code to programmatically generate SQL was challenging for data scientists, especially those with academic backgrounds, since they were mostly familiar with R and SQL. At the time, the only Python SQL parser was [sqlparse]([https://github.com/andialbrecht/sqlparse), which is not actually a parser but a tokenizer, so having users write raw SQL into the platform wasn't really an option. Some time later, I randomly stumbled across [Crafting Interpreters](https://craftinginterpreters.com/) and realized that I could use it as a guide towards creating my own SQL parser/transpiler.
 
 Why did I do this? Isn't a Python SQL engine going to be extremely slow?
 
 The main reason why I ended up building a SQL engine was...just for **entertainment**. It's been fun learning about all the things required to actually run a SQL query, and seeing it actually work is extremely rewarding. Before SQLGlot, I had zero experience with lexers, parsers, or compilers.
 
-In terms of practical use cases, I planned to use the Python SQL engine for unit testing SQL pipelines. Big data pipelines are tough to test because many of the engines are not open source and cannot be run locally. With SQLGlot, you can take a SQL query targeting a warehouse such as [Snowflake](https://www.snowflake.com/en/) and seamlessly run it in CI on mock Python data. It's easy to mock data and create arbitrary [UDFs](https://en.wikipedia.org/wiki/User-defined_function) because everything is just Python. Although the implementation is slow and unsuitable for large amounts of data (> 1 millon rows), there's very little overhead/startup and you can run queries on test data in a couple of milliseconds.
+In terms of practical use cases, I planned to use the Python SQL engine for unit testing SQL pipelines. Big data pipelines are tough to test because many of the engines are not open source and cannot be run locally. With SQLGlot, you can take a SQL query targeting a warehouse such as [Snowflake](https://www.snowflake.com/en/) and seamlessly run it in CI on mock Python data. It's easy to mock data and create arbitrary [UDFs](https://en.wikipedia.org/wiki/User-defined_function) because everything is just Python. Although the implementation is slow and unsuitable for large amounts of data (> 1 million rows), there's very little overhead/startup and you can run queries on test data in a couple of milliseconds.
 
 Finally, the components that have been built to support execution can be used as a **foundation** for a faster engine. I'm inspired by what [Apache Calcite](https://github.com/apache/calcite) has done for the JVM world. Even though Python is commonly used for data, there hasn't been a Calcite for Python. So, you could say that SQLGlot aims to be that framework. For example, it wouldn't take much work to replace the Python execution engine with numpy/pandas/arrow to become a respectably-performing query engine. The implementation would be able to leverage the parser, optimizer, and logical planner, only needing to implement physical execution. There is a lot of work in the Python ecosystem around high performance vectorized computation, which I think could benefit from a pure Python-based [AST](https://en.wikipedia.org/wiki/Abstract_syntax_tree)/[plan](https://en.wikipedia.org/wiki/Query_plan). Parsing and planning doesn't have to be fast when the bottleneck of running queries is processing terabytes of data. So, having a Python-based ecosystem around SQL is beneficial given the ease of development in Python, despite not having bare metal performance.
 
@@ -77,7 +77,7 @@ Once we have our AST, we can transform it into an equivalent query that produces
 
 1. It's easier to debug and [validate](https://github.com/tobymao/sqlglot/blob/main/tests/fixtures/optimizer) the optimizations when the input and output are both SQL.
 
-2. Rules can be applied a la carte to transform SQL into a more desireable form.
+2. Rules can be applied a la carte to transform SQL into a more desirable form.
 
 3. I wanted a way to generate 'canonical sql'. Having a canonical representation of SQL is useful for understanding if two queries are semantically equivalent (e.g. `SELECT 1 + 1` and  `SELECT 2`).