Update the sqlite3_analyzer utility program to provide more details about

the compression performance of ZIPVFS database files.

FossilOrigin-Name: fa5ed53296d45a5f963d20fb361555c432e0f31b

diff --git a/manifest b/manifest
index 01e580b938cb2cf409a7302674ad6cf83acc2f9b..a7c106eb239b3fba94f184797633fb2b7e9e7fb8 100644 (file)
--- a/manifest
+++ b/manifest
@@ -1,5 +1,5 @@
-C Fix\stest\scode\srelated\sto\sreporting\sthe\ssize\sof\soverflow\spages\sin\szipvfs\sdatabases.
-D 2011-10-05T17:36:27.323
+C Update\sthe\ssqlite3_analyzer\sutility\sprogram\sto\sprovide\smore\sdetails\sabout\nthe\scompression\sperformance\sof\sZIPVFS\sdatabase\sfiles.
+D 2011-10-05T18:18:13.395
 F Makefile.arm-wince-mingw32ce-gcc d6df77f1f48d690bd73162294bbba7f59507c72f
 F Makefile.in a162fe39e249b8ed4a65ee947c30152786cfe897
 F Makefile.linux-gcc 91d710bdc4998cb015f39edf3cb314ec4f4d7e23
@@ -956,7 +956,7 @@ F tool/showjournal.c b62cecaab86a4053d944c276bb5232e4d17ece02
 F tool/showwal.c f09e5a80a293919290ec85a6a37c85a5ddcf37d9
 F tool/soak1.tcl 8d407956e1a45b485a8e072470a3e629a27037fe
 F tool/space_used.tcl f714c41a59e326b8b9042f415b628b561bafa06b
-F tool/spaceanal.tcl 58d357384760020443c41d01db612c9809c75de7
+F tool/spaceanal.tcl 7ba8b9784fe7c4fb89d3d9ca012d41c9d74b3c95
 F tool/speedtest.tcl 06c76698485ccf597b9e7dbb1ac70706eb873355
 F tool/speedtest16.c c8a9c793df96db7e4933f0852abb7a03d48f2e81
 F tool/speedtest2.tcl ee2149167303ba8e95af97873c575c3e0fab58ff
@@ -967,7 +967,7 @@ F tool/symbols.sh caaf6ccc7300fd43353318b44524853e222557d5
 F tool/tostr.awk e75472c2f98dd76e06b8c9c1367f4ab07e122d06
 F tool/vdbe-compress.tcl d70ea6d8a19e3571d7ab8c9b75cba86d1173ff0f
 F tool/warnings.sh fbc018d67fd7395f440c28f33ef0f94420226381
-P 328cc1867ffbbf1c953dfd843649f5f209c8e6ec
-R 9966c1a1728c681681f9fe1b1d6773fc
-U dan
-Z 69c3b890ba0b067ca386e3a142264668
+P ad7c9eed8bbd607babce4f5965f587c873e7bc02
+R e170bea645862be3dfb0e179487437ed
+U drh
+Z 9854118ea63abc101ac650ae8a3b25fa

diff --git a/manifest.uuid b/manifest.uuid
index a41affd0c68cac794279d242981b4ee22a3c1431..c7274fa7180d54176ef02d4e75a57d21c9ae3838 100644 (file)
--- a/manifest.uuid
+++ b/manifest.uuid
@@ -1 +1 @@
-ad7c9eed8bbd607babce4f5965f587c873e7bc02
\ No newline at end of file
+fa5ed53296d45a5f963d20fb361555c432e0f31b
\ No newline at end of file

diff --git a/tool/spaceanal.tcl b/tool/spaceanal.tcl
index ba7d1635b3e0790891f486d8ee06d9380193caf5..22c63f3fbc9b31130df8e557646ab6386f6365bc 100644 (file)
--- a/tool/spaceanal.tcl
+++ b/tool/spaceanal.tcl
@@ -4,18 +4,29 @@
 #
 
 if {[catch {
-if {![info exists argv0]} {
-  set argv0 [file rootname [file tail [info nameofexecutable]]]
-}
-
 # Get the name of the database to analyze
 #
-#set argv $argv0
-if {![info exists argv] || [llength $argv]!=1} {
+proc usage {} {
+  set argv0 [file rootname [file tail [info nameofexecutable]]]
   puts stderr "Usage: $argv0 database-name"
   exit 1
 }
-set file_to_analyze [lindex $argv 0]
+set file_to_analyze {}
+set flags(-pageinfo) 0
+append argv {}
+foreach arg $argv {
+  if {[regexp {^-+pageinfo$} $arg]} {
+    set flags(-pageinfo) 1
+  } elseif {[regexp {^-} $arg]} {
+    puts stderr "Unknown option: $arg"
+    usage
+  } elseif {$file_to_analyze!=""} {
+    usage
+  } else {
+    set file_to_analyze $arg
+  }
+}
+if {$file_to_analyze==""} usage
 if {![file exists $file_to_analyze]} {
   puts stderr "No such file: $file_to_analyze"
   exit 1
@@ -24,7 +35,8 @@ if {![file readable $file_to_analyze]} {
   puts stderr "File is not readable: $file_to_analyze"
   exit 1
 }
-if {[file size $file_to_analyze]<512} {
+set true_file_size [file size $file_to_analyze]
+if {$true_file_size<512} {
   puts stderr "Empty or malformed database: $file_to_analyze"
   exit 1
 }
@@ -85,6 +97,8 @@ db func isleaf isleaf
 db func isinternal isinternal
 db func isoverflow isoverflow
 
+set isCompressed 0
+set compressOverhead 0
 set sql { SELECT name, tbl_name FROM sqlite_master WHERE rootpage>0 }
 foreach {name tblname} [concat sqlite_master sqlite_master [db eval $sql]] {
 
@@ -107,6 +121,13 @@ foreach {name tblname} [concat sqlite_master sqlite_master [db eval $sql]] {
     FROM temp.dbstat WHERE name = $name
   } break
 
+  set total_pages [expr {$leaf_pages+$int_pages+$ovfl_pages}]
+  set storage [expr {$total_pages*$pageSize}]
+  if {!$isCompressed && $storage>$compressed_size} {
+    set isCompressed 1
+    set compressOverhead 14
+  }
+
   # Column 'gap_cnt' is set to the number of non-contiguous entries in the
   # list of pages visited if the b-tree structure is traversed in a top-down
   # fashion (each node visited before its child-tree is passed). Any overflow
@@ -203,7 +224,7 @@ proc divide {num denom} {
 # the $where clause determines which subset to analyze.
 #
 proc subreport {title where} {
-  global pageSize file_pgcnt
+  global pageSize file_pgcnt compressOverhead
 
   # Query the in-memory database for the sum of various statistics 
   # for the subset of tables/indices identified by the WHERE clause in
@@ -279,6 +300,7 @@ proc subreport {title where} {
   statline {Number of entries} $nleaf
   statline {Bytes of storage consumed} $storage
   if {$compressed_size!=$storage} {
+    set compressed_size [expr {$compressed_size+$compressOverhead*$total_pages}]
     set pct [expr {$compressed_size*100.0/$storage}]
     set pct [format {%5.1f%%} $pct]
     statline {Bytes used after compression} $compressed_size $pct
@@ -419,7 +441,13 @@ statline {Number of tables in the database} $ntable
 statline {Number of indices} $nindex
 statline {Number of named indices} $nmanindex
 statline {Automatically generated indices} $nautoindex
-statline {Size of the file in bytes} $file_bytes
+if {$isCompressed} {
+  statline {Size of uncompressed content in bytes} $file_bytes
+  set efficiency [percent $true_file_size $file_bytes]
+  statline {Size of compressed file on disk} $true_file_size $efficiency
+} else {
+  statline {Size of the file in bytes} $file_bytes
+}
 statline {Bytes of user payload stored} $user_payload $user_percent
 
 # Output table rankings
@@ -432,6 +460,24 @@ mem eval {SELECT tblname, count(*) AS cnt,
           FROM space_used GROUP BY tblname ORDER BY size+0 DESC, tblname} {} {
   statline [string toupper $tblname] $size [percent $size $file_pgcnt]
 }
+if {$isCompressed} {
+  puts ""
+  puts "*** Bytes of disk space used after compression ***********************"
+  puts ""
+  set csum 0
+  mem eval {SELECT tblname,
+                  int(sum(compressed_size)) +
+                         $compressOverhead*sum(int_pages+leaf_pages+ovfl_pages)
+                        AS csize
+          FROM space_used GROUP BY tblname ORDER BY csize+0 DESC, tblname} {} {
+    incr csum $csize
+    statline [string toupper $tblname] $csize [percent $csize $true_file_size]
+  }
+  set overhead [expr {$true_file_size - $csum}]
+  if {$overhead>0} {
+    statline {Header and free space} $overhead [percent $overhead $true_file_size]
+  }
+}
 
 # Output subreports
 #