chiark
/
gitweb
/
~ian
/
topbloke-formulae.git
/ commitdiff
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
| commitdiff |
tree
raw
|
patch
|
inline
| side by side (parent:
ebe387b
)
strategy: wip
author
Ian Jackson
<ijackson@chiark.greenend.org.uk>
Wed, 25 Apr 2012 22:06:25 +0000
(23:06 +0100)
committer
Ian Jackson
<ijackson@chiark.greenend.org.uk>
Wed, 25 Apr 2012 22:06:25 +0000
(23:06 +0100)
strategy.tex
patch
|
blob
|
history
diff --git
a/strategy.tex
b/strategy.tex
index 3760941afd1c8b4707132d6b97cbdf9d6a155d61..8c87f918141929d2c4bf36ff28ebc32b14a19ffe 100644
(file)
--- a/
strategy.tex
+++ b/
strategy.tex
@@
-1,42
+1,59
@@
\section{Strategy}
\section{Strategy}
-We start with some commits $S_0 \ldots S_n$
-(where $S_0 = L$ and is the current git ref for $\pl$).
+When we are trying to do a merge of some kind, in general,
+we want to merge some commits $S_0 \ldots S_n$.
+We'll write $S_0 = L$. We require that $L$ is the current git ref
+for $\patchof{L}$.
%Let $\set E_{\pc} = \bigcup_i \pendsof{S_i}{\pc}$.
%Let $\set E_{\pc} = \bigcup_i \pendsof{S_i}{\pc}$.
-Invoke Plan $\patchof \pl$ where the algorithm Plan $\pc$ is as
-follows:
+\subsection{Notation}
+\begin{basedescript}{
+\desclabelwidth{5em}
+\desclabelstyle{\nextlinelabel}
+}
+\item[ $\depsreqof{K}$ ]
+The set of direct dependencies (in the form $\py$)
+requested in the commit $K$ ($K \in \pn$) for the patch $\p$.
-Notation:
+\item[ $\pc \succ_1 \{ \p, \pq \ldots \}$ ]
+The Topbloke commit set $\pc$ has as direct contributors
+(see below) exactly $\p, \pq, \ldots$. This is an acyclic relation.
- $\pc \succ_1 \{ \p, \pq \ldots \}$
- the Topbloke commit set $py$ has as direct contributors exactly
- $\p, \pq, \ldots$. This is an acyclic relation.
+\item[ $\p \succ \pq$ ]
+The commit set $\p$ has as direct or indirect contributor the commit
+set $\pq$.
+This is an acyclic relation, and is the completion of $\succ_1$ into a
+partial order.
- Extend this into the partial order $\succ$.
+\end{basedescript}
-$\py \succ \pq$
+\subsection{Planning phase}
+We use a recursive planning algorith, recursing over Topbloke commit
+sets (ie, sets $\py$ or $\pn$). We'll call the commit set we're
+processing at each step $\pc$. We start with $\pc = \pl$
+where $\pl = \patchof{L}$.
-We intend to merge all $\set E_{\pc} = \{ E_{\pc,j \ldots} \}$
+At each recursive step
+we intend to merge all $\set E_{\pc} = \{ E_{\pc,j \ldots} \}$
and all the direct contributors of $\pc$ (as determined below)
into the existing git ref for $\pc$, to make $T_{\pc}$.
and all the direct contributors of $\pc$ (as determined below)
into the existing git ref for $\pc$, to make $T_{\pc}$.
-The direct contributors of $\pcn$ are the
t
opbloke commit sets
+The direct contributors of $\pcn$ are the
T
opbloke commit sets
corresponding to the tip branches for the direct dependencies of
$\pc$.
The sole direct contributor of $\pcy$ is $\pcn$.
corresponding to the tip branches for the direct dependencies of
$\pc$.
The sole direct contributor of $\pcy$ is $\pcn$.
+\subsubsection{Planning step for $\pc = \pcn$.}
-For $\pc = \pcn$, choose an (arbitrary, but ideally somehow optimal in
+FIXME DEFINE $\set E$
+
+Choose an (arbitrary, but ideally somehow optimal in
a way not discussed here) ordering of $\set E_{\pc}$, $E_j$ (for
$j = 1 \ldots m$). Remove from that set (and ordering) any $E_j$ which
are $\le$ and $\neq$ some other $E_k$.
a way not discussed here) ordering of $\set E_{\pc}$, $E_j$ (for
$j = 1 \ldots m$). Remove from that set (and ordering) any $E_j$ which
are $\le$ and $\neq$ some other $E_k$.
-Notation: write $\depsreqof{K}$ to mean the direct dependencies
-(in the form $\py$) requested in the commit $K$.
-
Initially let $T_{\pc,0}$ be the git ref for $\pcn$. And let
$\set D_0 = \depsreqof{T_{\pc,0}}$.
For each $E_j$ starting with $j=1$ choose a corresponding intended
Initially let $T_{\pc,0}$ be the git ref for $\pcn$. And let
$\set D_0 = \depsreqof{T_{\pc,0}}$.
For each $E_j$ starting with $j=1$ choose a corresponding intended