Updates

Author: eoinkelly

databases/LearningSQLExample.sql renamed to databases/book-leaning-sql-code/LearningSQLExample.sql

@@ -0,0 +1,83 @@
+# Character sets
+
+A character set defines 3 things
+
+1. a set of glyphs
+2. an integer representation of each glyph
+3. a way of encoding those integers as bytes
+
+ASCII is a simple character set
+
+* defines a set of glyphs
+* defines the integer that each glyph is represented by
+* defines how to represent that integer in bytes
+
+###  Collation
+
+* A character set needs some rules about how to compare characters in the set
+  e.g. for sorting and string comparisons e.g. less than, equal, greater than
+  etc.
+* This set of rules is called a collation
+* A character set always has at least one collations
+* A character set may have more than one collation
+* => A collation is a property of a character set and has no meaning outside
+  that character set.
+
+Examples of collations
+
+* binary
+    * compare characters based on the integers they are encoded as
+    * very simplistic
+* case-insensitve
+    * treat upper and lower case letters as equal
+* case sensitive
+
+Common encoding names have a pattern:
+
+    <character set name>_<language_name>_<suffix>
+
+where `<suffix>` can be:
+
+    cs => case sensitivk
+    ci => case insensitive
+    bin => binary
+
+* A collation is tied to a character set encoding
+    * the same collation name may exist in multiple character sets
+* There are 3 collations available on all platforms
+    1. `default`
+    2. `C`
+    3. `POSIX`
+* C and POSIX have identical behaviours - they both specify "traditional C" behaviour
+    * only uppercase A to Z considered "letters"
+    * sorting done strictly by character code byte value
+    * postgres considers them different collations so you can't compare a a
+      column with a "C" locale and a column with a "POSIX" locale.
+
+# Unicode includes a character set, collations and encodings
+
+* Unicode includes the "Universal character set standard"
+    * includes 120k+ characters in almost all the worlds languages
+* Unicode v8.0 was released July 2015
+* Unicode assigns each glyph a "code point" which is a positive integer
+* Each code point needs to be represented as a binary number - _how_ that
+  happens is specified by the "encoding"
+* Unicode has a number of possible encodings e.g. UTF-8, UTF-16
+* UTF-8 is an _encoding_ i.e. a way of turning a list of codepoints integers into bytes
+* UTF-8 uses 1 byte for a codepoint if it can and 2-4 bytes if necessary
+    * => it is a variable length encoding: 1 to 4 bytes
+    * It is a superset of ASCII
+
+* Unicode defines a _customisable_ collation algorithm
+    * assigns a tuple of weighting floats to each glyph
+    * it has a table with a default collation for _all_ unicode chars
+        * the Default Unicode Collation Element Table (DUCET).
+        * customizable for different languages
+
+# Examples of character sets
+
+* US-ASCII
+* ISO-8859-1 aka latin1
+* Unicode
+
+Others http://www.iana.org/assignments/character-sets/character-sets.xhtml